University of Wisconsin Milwaukee
UWM Digital Commons
Theses and Dissertations
5-1-2014
A New Paradigm for Punctuation
Albert Edward Krahn
University of Wisconsin-Milwaukee
Follow this and additional works at: http://dc.uwm.edu/etd
Part of the Linguistics Commons, and the Modern Languages Commons
Recommended Citation
Krahn, Albert Edward, "A New Paradigm for Punctuation" (2014). Theses and Dissertations. Paper 465.
This Dissertation is brought to you for free and open access by UWM Digital Commons. It has been accepted for inclusion in Theses and Dissertations
by an authorized administrator of UWM Digital Commons. For more information, please contact [email protected].
A NEW PARADIGM FOR PUNCTUATION
by
Albert E. Krahn
A Dissertation Submitted in
Partial Fulfillment of the
Requirements for the Degree of
Doctor of Philosophy
in Linguistics
at
The University of Wisconsin-Milwaukee
May 2014
ABSTRACT
A NEW PARADIGM FOR PUNCTUATION
by
Albert E. Krahn
The University of Wisconsin-Milwaukee, 2014
Under the Supervision of Professor Fred R. Eckman
This is a comprehensive study of punctuation, particularly the uses to
which it has been put as writing developed over the centuries and as it
gradually evolved from an aid to oral delivery to its use in texts that were
read silently. The sudden need for standardization of punctuation which
occurred with the start of printing spawned some small amount of interest in
determining its purpose, but most works after printing began were devoted
mainly to helping people use punctuation rather than try to discover why it
was being used. Gradually, two main views on its purpose developed: it was
being used for rhetorical purposes or it was needed to reveal the grammar in
writing. These views are still somewhat in place.
ii
The community of linguists took little notice of writing until the last
few centuries and even less notice of punctuation. The result was that few
studies were done on the underlying purpose for punctuation until the
twentieth century, and even those were few and far between, most of them
occurring only in the last thirty years.
This study argues that neither rhetoric nor grammar is directly the
basis for punctuation. Rather, it responds to a schema that determines the
order of the words in spoken and written English, and it is a linguistic
concept without question. The special uses of the features of punctuation are
discussed, as well as some anomalies in its use, some ideas for more studies,
and some ideas for improving the teaching of punctuation.
iii
© Copyright by Albert E. Krahn, 2014
All Rights Reserved
iv
TABLE OF CONTENTS
CHAPTER
1
Introduction
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
2
PAGE
The Purpose of the Study ...............................................1
Definitions in Textbooks and Handbooks ....................5
Punctuation in Reference Works ..................................8
Some Scholarly Approaches to Punctuation ...............10
Defining Language .......................................................16
Defining Writing .........................................................16
The Symbols found in Writing .....................................20
Summary .......................................................................23
The Evolution of Punctuation
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
2.10
2.11
2.12
2.13
2.14
2.15
2.16
2.17
2.18
2.19
2.20
2.21
2.22
Introduction ..................................................................24
The Earliest Punctuation ..............................................25
Writing as a Technology ................................................26
Problems with Terminology .........................................27
The Genre Confusion ....................................................33
Characteristics of the Evolution ..................................34
The Greek Punctuation ................................................35
Punctuation and the Roman Empire ...........................38
The Glory of Roman Punctuation ................................41
The Empire after the Fall .............................................45
A Small Change with a Big Effect ...............................47
The Early Middle Ages .................................................48
The Role of Important Individuals in the Middle
Ages ................................................................................52
The Ninth Century Turnaround ...................................54
The Great Division ........................................................57
The Humanists of the Fourteenth Century .................58
The Modes of Medieval Punctuation ...........................59
The Impact of Printing on Punctuation ......................61
Syntax vs. Rhetoric: The Theoretical Argument ........63
Punctuation in Concert with Music ............................67
The Triumph of Technology ........................................69
Conclusion ....................................................................74
v
3
Punctuation Scholarship in the Twentieth Century
3.1
3.2
Introduction ..................................................................75
A Century of Little Progress ........................................76
3.3
Published Works on Punctuation in the 1900s and early 2000
3.3.1 Husband, Punctuation: Its Principles
and Practices .....................................................77
3.3.2 Summey, Modern Punctuation .........................81
3.3.3 Summey, American Punctuation ......................84
3.3.4, Meyer, A Linguistic Study of American
Punctuation .......................................................87
3.3.5 Nunberg, The Linguistics of Punctuation ........89
3.3.6 Patt, Punctuation as a Means of
Medium-Dependent Presentation Structure
in English...........................................................92
3.4
Books Which Contain Some Linguistic Value
3.4.1 Treip, Milton’s Punctuation and
Changing English Usage, 1582-1676 ..............95
3.4.2 Lennard, But I Digress: The Exploitation
of Parentheses in English Printed Verse ..........95
3.4.3 Graham-White, Punctuation and Its
Dramatic Value inShakespearean Drama .......95
3.4.5 Saenger, Space Between Words:
The Origins of Silent Reading ..........................95
3.5
Dissertations Having a Linguistic Approach to Punctuation
3.5.1 Steegar, Prosody and Punctuation:
A Linguistic Approach .......................................96
3.5.2 Levinson, Punctuation and the Orthographic
Sentence: A Linguistic Analysis ........................97
3.5.3 Waller, The typographic contribution to
language: towards a model of typographic
genres and their underlying structures ...........97
3.5.4 Hill, A comma in parsing: a study
into the influence of punctuation
(commas) on contextually isolated
“garden-path” sentences ...................................97
vi
3.6
3.5.5 Jones, What’s the point? A (computational)
theory of punctuation .......................................98
3.5.6 Say, An information-based approach
to punctuation ..................................................98
3.5.7 Doran, Incorporating punctuation
into the sentence grammar: a lexicalized
tree adjoining grammar perspective ...............98
3.5.8 Vial, Take pause: musical punctuation
in the eighteenth century ..................................99
3.5.9 Grindlay, Missing the point: the effect
of punctuation on reading performance ........99
3.5.10 Schou, The problem of English punctuation ...99
3.5.11 Arriola-Nickell, Punctuation as symbol:
experiencing archetypal patterns through
personal narrative ...........................................100
Conclusion ..................................................................100
Chapter 4 The New Paradigm for Punctuation
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
4.10
4.11
4.12
4.13
4.14
4.15
4.16
4.17
4.18
4.19
4.20
4.21
Introduction: Content and Form ...............................102
Hypotheses for Form ..................................................103
The Sentence ..............................................................104
The Sentence by Parts ...............................................108
The Parsing Process ...................................................110
Sentences without Meaning ......................................115
The Schema ................................................................116
Yaggy and the Canonical Sentence ..........................121
The New Paradigm ....................................................123
English Writing as a Graphic ...................................124
Gestalt and the Text ...................................................130
Locating Punctuation ................................................136
The Sample Paragraph with Punctuation ...............139
The Sample Paragraph without Punctuation ..........139
The Importance of Space ...........................................140
Putting Spaces back into the Sample Paragraph ....145
Phonological Punctuation ..........................................147
Morphological Punctuation .......................................147
The Sample Paragraph with only Phonological
and Morphological Punctuation now added ............ 149
Static Attention Devices ............................................149
Punctuation and Markedness ...................................150
vii
4.22
4.23
4.24
4.25
4.26
4.27
4.28
Phonemes and Graphemes..........................................157
Punctuating the Dynamic Axis: The Comma.............160
The Role of Interpolations...........................................169
The Role of Adjuncts....................................................170
The Role of Terminal Punctuation..............................170
The Absorbtion Rule ..................................................170
Summary ....................................................................171
Chapter 5 An In-depth look at some Punctuation Features;
Anomalies; the Puncteme; Research and Teaching Ideas
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
5.10
5.11
5.12
5.13
5.14
5.15
5.16
5.17
5.18
5.19
Introduction: Speaking and Writing ........................173
The Comma as a Tool ................................................174
Recursion and the Comma .......................................177
Interpolations ............................................................178
Parenthesis and Quotation .......................................180
Diacritics ....................................................................184
The Apostrophe ..........................................................186
The Hyphen, Ellipsis, and the Virgule ....................189
Capitals and Terminal Punctuation ........................190
The Adjuncts: The Semicolon and the Colon ..........193
Graphemes and Punctemes .......................................198
Punctuation and Research .........................................200
Manuscripts and Punctuation ...................................202
The Forensic Study of Writing ..................................202
Punctuation for New Written Languages ................203
Punctuation and the Future of English ....................204
Teaching Punctuation ................................................205
A Teaching Method for Punctuation .........................209
Summary ....................................................................213
Chapter 6 Postscript
6.1
6.2
6.3
6.4
Some Things Don’t Change and Some Might ...........214
Punctuation: Sight or Sound? ....................................214
The Cross-Linguistic Study of Punctuation ..............215
Conclusion ...................................................................217
References ......................................................................................218
viii
1
Chapter 1
Introduction
1.1 The Purpose of the Study
The goal of this study is to look at the current beliefs about punctuation, how
it evolved to its current state, what scholarship regarding punctuation has
taken place over the past century, and look at a new way to consider how
punctuation functions. The particular behavior of some individual features
are also examined, along with some possibilities for research and teaching.
Chapter 1 will survey some of the more contemporary attitudes toward
punctuation and define some of the terms that will be discussed later.
Chapter 2 reviews the stages in the development of punctuation,
dwelling not on individual symbols but on the larger ideas that were
motivating the attempts to use them.
Chapter 3 is a survey of the literature on punctuation from the past
century, the period of time during which the punctuation we use today should
have been coming under more scrutiny because of the large quantity of
writing being published.
Chapter 4 will develop the hypothesis and explore how the form of
language rather than the content is the structure that punctuation is really
functioning in. It is what the purpose is, rather than how it appears to be
2
doing it, that is of importance here in the hope of proving the thesis:
Punctuation is a system of symbols and graphic features used to
protect the integrity of the sentence in the English writing system
and make it visible.
Chapter 5 examines some individual punctuation features that
scholars have found of particular interest because of what they can do, some
anomalies that exist in the present system, along with some ideas for future
research and the teaching of punctuation.
It seems appropriate that William of Ockham was born in Surrey,
England (in 1285), around the time that silent reading was becoming the
norm in libraries, because this study is going to use a principle that is
associated with his name, Ockham’s Razor. Briefly, it reads like this:
One should always choose the simplest explanation of a
phenomenon, one that requires the fewest leaps of logic.
(Beckett, 1994)
This study aims to adopt Ockham’s approach and instead seek the more
general “cause,” the reason why there is a need for punctuation and the
principle involved. This will involve looking at it from a number of points of
view.
The many ways people have tried to determine the function of
punctuation to date reminds one of the story of the elephant and the blind
men. Each one examines a different part of the elephant and each makes a
3
different judgment about its reality. Punctuation is very much the same.
Many people have studied and commented on punctuation, and each seems to
see a different phenomenon. They tend to look at what is on a page already
and concentrate on the “effects” of punctuation, usually in local places. The
concern here is to look at contemporary English and how punctuation works
as a linguistic feature in the English orthography.
Studies of punctuation take a variety of approaches, including its
historic development, the frequency of use of certain symbols, the special and
varied uses of individual symbols, how it impacts the processing of English in
a corpus, its use as a stylistic device, how it does or does not duplicate the
sounds of speech, whether it is related to logic or rhetoric or prosody or just
something to help those who want to read a text out loud, and many more.
Greta Little stated what still needed to be done yet in studies of
punctuation back in 1986 in an article titled “Punctuation as a linguistic
phenomena”:
Like other linguistic phenomena, punctuation must be examined
in a variety of ways. We need to know its history, how it evolved
through time . . . . We need to establish the various ways in
which the punctuation symbols are able to convey meaning—
how they function.
(Little, 1986:71-72.)
Unfortunately, most of the studies since she wrote have not accomplished
what she was looking for. Only one of Little’s wishes has been granted: there
4
have been a number of good studies on the history of punctuation. To make
matters worse, the linguistic community has not been very sympathetic to
Little’s invocation. Not only has punctuation not gathered much interest to
linguists, but even writing itself has gotten a cold shoulder, perhaps partly
because of what was said by some significant scholars in the last century.
About 80 years ago, Leonard Bloomfield, who spent some time in
Wisconsin during his growing up and college years, wrote that “Writing is not
language, but merely a way of recording language by visible marks”
(Bloomfield, 1933:21). Edward Sapir, an important contemporary of
Bloomfield, apparently accepted writing as “language” but had little to say
about it for this reason:
Written language is thus a point-to-point equivalence, to borrow
a mathematical phrase, to its spoken counterpart. The written
forms are secondary symbols of the spoken ones—symbols of
symbols—yet so close in the correspondence that they may . . .
be entirely substituted for the spoken ones.
(Sapir,1921:20.)
Sapir’s only reference to punctuation is in his Preface to the book where this
is stated, where he claims that “There is not a single diacritical mark in the
book” (vi). Both Bloomfield and Sapir apparently inherited Saussure’s
approach to writing because he had made it quite clear where he stood:
5
Language and writing are two distinct systems of signs; the
second exists for the sole purpose of representing the first. The
linguistic object is not both the written and the spoken forms of
words; the spoken forms alone constitute the object.
(Bally et al., 1959:23-24.)
Apparently, the result of these statements was that, for a good part of the
twentieth century, writing was not considered that important to linguists.
Along with this view of writing, punctuation was of even less concern to
linguists and apparently still is not considered that important. You are
unlikely to find even a mention of punctuation in most books related to
linguistics, much less a definition of it. Definitions of punctuation are also
difficult to find in books and articles on related subjects as well, regardless of
their level.
1.2 Definitions in Textbooks and Handbooks
Students looking for assistance in understanding punctuation as a concept
get very little direction. Most of the definitions of punctuation in writing texts
are not helpful. Many writing texts no longer even offer a definition or
description for punctuation but go right to the “rules.” Those that have
offered descriptions of punctuation over the years have tended to use
metaphors, make reference to speech characteristics, or merely offer general
6
or vague descriptions. Here are some typical examples of definitions that
have made it into textbooks over the years:
Punctuation is to writing what notation is to music: it allows
the eye to re-create from the page the sounds the author of the
composition had in mind. Both are necessary and exacting
systems. Just as musicians know the crucial difference between
a quarter note and a half note, so writers know the crucial
difference between a comma and a semicolon, between brackets
and parentheses.
(Mulderig and Elsbree, 1990:605.)
Most punctuation marks represent the pauses and stops we
would use in speaking. Periods, question marks, exclamation
points, semicolons, and colons are stop marks—the “red lights”
of writing. Commas and dashes are pause marks—the “amber
lights” that tell us to slow down momentarily. Some punctuation
marks separate words and ideas; others group and keep together
related ideas; still others set aside words for special emphasis.
(Ellsworth, 1985:12.)
Punctuation consists of cue marks for joining and separating
words, phrases, clauses, and sentences. The purpose of
punctuation is to clarify what otherwise would seem vague or
confusing to a reader.
(Kuiper and Luke, 1992:403.)
7
Punctuation is designed to mirror the way we speak. For
example, a period is supposed to reflect the amount of time (a
period) that it takes to say one sentence. If you were to read a
document out loud, the periods would signal places to breathe.
Similarly, commas are used to signal a pause. When you come
across a comma in a sentence, you pause slightly.
(Johnston-Sheehan, 2005:275.)
. . . by punctuation, the writer can suggest what the speaker can
reveal with gesture, pause, tone, and pitch. Or, to use the terms
of many modern grammarians, punctuation is to writing what
supra-segmental phonemes . . . are to the spoken sentence.
Imaginative use of punctuation can strengthen prose;
conventional use of punctuation, according to codes refined by
printers and editors, promotes accuracy and clarity.
(Gorrell and Laird, 1967:467.)
These are just a few of the many different reasons for punctuation that one
can find in handbooks and writing textbooks.
The many reference books devoted directly to helping people use
punctuation that are on my shelves are not much better. Most have no
definition of punctuation at all. Some offer metaphors similar to what the
writing textbooks offer. Lynne Truss, in her Eats, Shoots & Leaves, a bestseller a few years ago, talks about defining punctuation but never really gets
8
around to doing it. The only book that even comes close is Eric Partridge’s
You Have a Point There. Partridge begins with comments on punctuation
from the writings of sixteen people who wrote on the subject between 1640
and 1943 (Partridge, 1953:3-6). They resemble the definitions found in the
texts and handbooks above, and Partridge never offers one of his own. Only
one handbook makes a stated offering of a definition, but it is clearly only
partly serious:
Punctuation: (1) A bunch of impossible-to-figure-out marks,
invented by the devil to give writers a foretaste of hell, taught in
a hundred confusing and contradictory ways. (2) A code, used in
writing, that is often necessary for meaning and emphasis. The
code originated in attempts to capture, in text, the various stops,
pauses, and inflections of speech. Today it is logical in
application.
(Lauchman, 2010:17.)
Lauchman’s first definition probably sums up the situation quite well.
Generally, the definitions you do find in textbooks, handbooks, and books on
punctuation are certainly not helpful, despite the many “rules” that they
offer.
1.3 Punctuation in Reference Works
Of the twelve dictionaries and encyclopedias devoted exclusively to linguistics
on the shelf in my study, only three have any entry at all for punctuation.
9
Two of them devote one sentence to it. The third devotes about a page to a
brief history of punctuation. A random check of ten grammars of different
languages in the library at the University of Wisconsin-Milwaukee revealed
that only two of them made any mention at all of punctuation, which
consisted of an index listing and a few pages in one of them. The
International Encyclopedia of Linguistics, a huge 2,218 page work, devotes
less than two pages to punctuation (Frawley, 2003). Also, The Writing
Systems of the World, a discussion of many languages at 922 pages, has a
dozen entries for punctuation which might add up to a total of about three or
four pages, but most of them relate only to diacritics (Daniels and Bright,
1996). The Chicago Manual of Style (Sixteenth Edition, 2010), which
otherwise offers very good advice about how to use punctuation, never
bothers to define it. A Comprehensive Grammar of the English Language
(Quirk, et al. 1985), a work which is often cited in discussions of punctuation,
states that punctuation serves two purposes, separation and specification,
and sees the features of punctuation as forming a hierarchy, but does
otherwise list a number of reasons for the way punctuation works:
The punctuation mark specifies a grammatical, semantic, or
pragmatic function, sometimes in addition to the marking of
separation. [So] punctuation practice is governed primarily by
grammatical considerations and is related to grammatical
distinctions. Sometimes it is linked to intonation, stress,
10
rhythm, pause, or any other of the prosodic features which
convey distinctions in speech, but the link is neither simple nor
systematic, and traditional attempts to relate punctuation
directly to (in particular) pauses are misguided.
(Quirk et al, 1985:1610-1612.)
There is no attempt here to locate a single principle that underlies the need
for punctuation. Instead of seeking a single idea that might lurk behind all
punctuation features, they seem to just cobble together a number of the
current ideas.
1.4 Some Scholarly Approaches to Punctuation
One might suspect that scholars who have been reading and writing about
punctuation and studying it would come up with some new ways to approach
it, but they tend to keep looking for answers in the same places. The majority
of articles by scholars adhere to one of the two approaches as the source of
punctuation that are the most common: a relationship to speech features or a
relationship to grammar. William Chafe, who has written a number of
articles on punctuation, insists that his approach is correct:.
At this point some readers may object that the signaling of
prosody is only one of the functions of punctuation, and perhaps
not the primary one. Although that is a common belief, and
`
although certainly there are instances of punctuation that do not
11
serve prosodic ends, I will defend the position here that those
instances are departures from its main function, which is to tell
us something about a writer’s intentions with regard to prosody
of that inner voice.
(Chafe, 1988:397.)
Fernando Poyatos has taken an even broader stand and places punctuation
in the category of non-verbal communication, a study he is immersed in:
. . . although punctuation reveals a conscious effort to symbolize
speech for the better evocation of its semantic variations and the
avoidance of too conspicuous ambiguities, it simultaneously, and
quite unwittingly, evokes and marks the co-occurrent body
movements and still positions that are an integral part of the
kinetic-acoustic continuum of human and animal
communication.
(Poyatos, 1981:91.)
Poyatos has produced many large charts in which he shows relationships
between what occurs in speech situations with what seems to occur in
punctuation in writing. Poyatos also believes that we have insufficient items
for showing in writing all the things that occur in speech and has invented a
number of new symbols to help writers elaborate their texts with a collection
of new punctuation features, something which others have also been doing
over the years
The belief that grammar is the model for punctuation is firmly
established in an article by Karsten Schou. After reviewing some historical
12
documents, he comes to this conclusion:
We have now seen how the discussion of the history of English
punctuation theory centres on the role that syntax has played.
The general experience is that syntax has been central at least
since 1600, although prosody played and still plays a certain
part. Punctuation and its theory have moved towards an
increasingly syntactic orientation. By examining sources in
punctuation theory we have seen that in the seventeenth
century punctuation was grammatical in the sense that it was
frequently used to mark boundaries of syntactic units, mainly
based on criteria of form.
(Schou, 2007:213.)
In other words, grammar is in charge, but prosody still gets a mention. But,
strangely, the word “form” turns up at the end of his statement.
Furthermore, what is fascinating about all the discussions of
punctuation and grammar is that almost none of them mention which
grammar they are talking about. It is spoken of as some kind of universal
system that everyone knows. This occurs in spite of the fact that dozens of
grammars have been proposed in the last century alone. One reference work,
Concise Encyclopedia of Syntactic Theories, published in 1996, lists and
describes at least thirty different grammars, and it is possible that some
more have been devised since then (Brown and Miller, 1996). They define
“grammar” as:
13
1. The study of language and the rules that govern its usage.
2. A description of the form of words and the manner in which
they combine to form phrases, clauses or sentences.
3. A systematic and explicit account of the structure of
(a) language according to the tenets of one or other of the
theories of modern linguistics.
(Brown and Miller, 1996:402.)
Based on these definitions, each grammar listed could approach the structure
of English—and hence the punctuation—in a different way, which certainly
would have some influence on how it is viewed and performs. Without some
extraordinary research, it would be difficult to determine which grammar
represents punctuation best. In the long run, it may not be necessary to do
that because punctuation functions in its own context, the written form of
English, a dialect which has its own requirements.
One of the few writers who has made a definite stand for a particular
grammar for punctuation is Christine D. Doran in her 1998 dissertation:
Punctuation marks are treated as full-fledged lexical items in a
Lexicalized Tree Adjoining Grammar, which is extremely wellsuited formalism for encoding punctuation in the sentence
grammar. Each mark anchors its own elementary trees and
imposes constraints on the surrounding lexical items.
(Doran, 1998:vi-vii.)
14
References to her approach occur in the literature, but not in everyday
discussions of grammar, probably because it is “a computational model of
punctuation” and as such has specific and limited uses.
Finally, there are even some scholars who find nothing useful in
punctuation and would dispense with it completely:
Punctuation conventions are culturally evolved aids to meaningmaking which members of a particular culture have in common,
and use as a resource for communicating through written
language. However, the particular set of conventions that
constitute the “rules” of punctuation for any culture are nothing
other than “conventions”, and are not based on any permanent
or universal principles of language. [So], while punctuation
conventions can be useful, the conventions themselves have no
intrinsic virtue and hence adhering to them should not be
treated as a marker of ability.
(Clark and Ivanic, 1997:205-207.)
This seems to give writers carte blanche to do whatever they like with
punctuation, a bit severe, it seems, but there is an undercurrent of this idea
in the writings of other scholars as well.
Were this study being written in the year 1014 CE instead of 2014 CE,
Scholes and Willis tell us we might have to include yet another purpose for
punctuation:
15
In its elocutionary function, punctuation serves as a set of
instructions for reading a text aloud. As such it is one aspect of
written speech . . . and a kind of phonetic transcription for
prosody (stress, pause, and intonation).
(Scholes and Willis, 1990:13.)
Certainly, different kinds of writing may use punctuation in somewhat
different ways. Also, it is obvious that punctuation involves conventions, but
the conventions evolve from some information that is important to the way
writing is presented. That is what we are looking for: a clue to what
punctuation does to language that makes it a sensible and necessary addition
to written English.
Harold Whitehall, although he subscribed to the idea that punctuation
was used to represent the sounds of speech, also said something prophetic
when he said that the “most important purpose” of punctuation is “to make
grammar graphic” (Whitehall, 1956:119).
Because of the many different approaches to punctuation, teaching it
to students is a significant challenge. Mina Shaughnessy, author of Errors &
Expectations, a book on teaching basic writing that shook up the English
teaching community, put it this way:
Unquestionably, one of the primary needs in literacy research
must be empirical investigation into the function of punctuation.
(Shaughnessy, 1977:81.)
16
Many years have passed since she wrote those words, and it really hasn’t
been accomplished yet.
1.5 Defining Language
At this point it might be appropriate to define some of the categories we will
be dealing with: language and writing. For the sake of expediency, I will use
Chomsky’s definition for “language”:
. . . I will consider a language to be a set (finite or infinite) of
sentences, each finite in length and constructed out of a
finite set of elements.
(Chomsky, 1966:13).
I am partial here to Chomsky’s definition because of his reference to
“sentences.” More will be said of sentences in a later chapter. The discussion
of punctuation will be involved with the concept of the sentence so it would be
beneficial to have on record some notion of the importance of the sentence as
a significant part of written language.
1.6 Defining Writing.
In many ways, punctuation is used to patch up an inadequate writing system.
You could probably say honestly that the more punctuation you encounter in
a writing system the more likely it is that the writing system is a poor
representative of the speech system. We are probably fortunate that English
17
is alphabetic, but as it evolved, it also has had to adapt in some ways that
have required punctuation and changed some of the conventions as the
language changed. Since this study is about the role of punctuation in the
English writing system, we should have some conception of what a writing
system is so we can get an idea of the kind of problems punctuation is trying
to solve.
The first warning we get about trying to define “writing” comes from
Florian Coulmas:
Every attempt at a single universal definition of writing runs
the risk of being either ad hoc or anachronistic, or informed by
cultural bias.
(Coulmas, 2003:2.)
Yet in an earlier work, he provided a rather reasonable list of probable
characteristics:
1. It consists or artificial graphical marks on a durable surface;
2. its purpose is to communicate something;
3. this purpose is achieved by virtue of the marks’ conventional
relation to language.
(Coulmas, 1991:17.)
Henry Rogers gets to the point quickly:
We can define writing as the use of graphic marks to represent
specific linguistic utterances.
(Rogers, 2005:2.)
Actually, Rogers is accomplishing two things at the same time here: he is in a
way also giving us a synonym for a quantity of speech: “specific linguistic
18
utterances.” Sooner or later, we may have to discuss the different forms in
writing. Utterance could be one of them: a stretch of written language which
has no other recognizable form, such as a text, a paragraph, a sentence, a
phrase, or a word. Utterance seems to be an abstract popular term for what
most of us would call simply speech.
I. J. Gelb, whose book on writing systems spawned many others, is also quite
brief:
Writing. A system of intercommunication by means of
conventional visible marks.
(Gelb, 1965:253.)
Michael D. Coe, author of a book on Mayan writing, gives a nod to speech:
Writing is speech put in visible form, in such a way that any
reader instructed in its conventions can reconstruct the vocal
messages.
(Coe, 1992:13.)
Geoffrey Sampson adds some synonyms to broaden out the definition a bit:
I shall use the terms script, writing system, or orthography, to
refer to a given set of written marks together with a particular
set of conventions for their uses.
(Sampson, 1985:19.)
John DeFrancis adds some information about the size for those who hadn’t
yet thought about the quantity it might take to be officially “writing”:
Full writing [as opposed to just a word or two on signs, for
example] is a system of graphic symbols that can be used to
convey any and all thought.
(DeFrancis, 1989:5.)
19
Peter T. Daniels, who edited The Writing Systems of the World with William
Bright, adds a new idea:
. . . writing is defined as a system of more less permanent marks
used to represent an utterance in such a way that it can be
recovered more or less exactly without the intervention of the
utterer.
(Daniels, 1996:3.)
Perhaps the longest definition comes from Floyd Lounsbury:
I speak of “writing” in its fullest sense in those instances where
we find graphic representation of complete sentences and the
concatenation of sentences into texts; but I also accept as
“writing,” though in a more attenuated sense, those instances
in which compound words and phrases are the maximum
attested units (as in representations of place names, personal
names, composite numerals, numerals with signs for things
tallied, etc.) but where the representation of the fully formed
sentences is not general.
(Lounsbury, 1989:203.)
Lounsbury, it seems, would accept the signs that DeFrancis refuses to accept.
John Mountford, who also talks about utterances, adds some things
that have so far been neglected:
Besides the verbal component in writing, we have a non-verbal
component. . . . [T]he non-verbal elements in writing can be
divided into ‘punctive’ (punctuation marks and other unit-
20
delimiting devices, including white space between sections,
indentation, word-space, etc.) and ‘parascriptal’ (italics,
asterisks, underlinings, changes of size, etc.).
(Mountford, 1980:229.)
The terms that get the most exposure in these descriptions are “graphical,”
“conventional,” and “utterance,” with a nod to “visible,” “system,” and
“sentence.” Some of the other ideas, such as “communicate,” “symbol,” and
“permanent,” are probably implied in some of the more common terms used.
Again, there is a term here that we may not have expected: “sentences.” Like
the word “form” that turned up earlier in a place where we might not expect
it, the word “sentence” here seems like a foreboding of something we might
find out about writing that we didn’t expect. But it would probably be
unnecessary to conduct a study to find out that, indeed, the sentence is not
only more common in writing than in speech but also more important. But we
might notice also that most of the terms used in the descriptions imply
something about “form.”
1.7 The Symbols found in Writing.
There are many different symbols found in writing, but not all of them are
punctuation. We need to sort out the different symbols that a reader might
encounter and determine which are probably punctuation and which are not.
There are at least three types:
21
1. Symbols that replace words: numbers, dollar signs,
ampersands, etc. Writers sometimes use one or the other.
They may turn up in a sentence but are usually understood
as a word rather than an abstract symbol representing some
non-verbal concept. Some symbols used in the sciences might
fit this category if they are pronounced.
2. Symbols which are extra-textual: bullet lists, paragraphing,
footnotes, etc. These are non-verbal and purely graphic. They
are not part of the text but rather the visual layout or design.
3. Symbols that have a linguistic function in a text: punctuation.
Some writers on writing systems and punctuation would like to consider
some of the items in 2. as part of a class called “macro-punctuation,” but they
are not considered so in this study. For our purposes, only those symbols
which have some influence on the other linguistic features in the writing will
be included. That list would include the following: comma, semicolon,
colon, period, question, exclamation, hyphen, dash, parenthesis,
brackets, apostrophe, diacriticals, capitals, bold, underline, italics,
ellipsis, and space. Some features that are also possible might include a
change of font size or appearance, which would be functioning in the same
way as bold, for example. The virgule seems to be feature in transition. It
turns up on occasion in expressions such as “and/or,” but it doesn’t seem to
22
have achieved tenure yet. Its influence on the linguistic features seems
questionable.
The above features are to be found in four major sources: A
Comprehensive Grammar of the English Language (1985), The Oxford
English Grammar (1996), The Chicago Manual of Style (2010), The
Cambridge Encyclopedia of the English Language (1995). There are only
some minor disagreements among these sources. It is interesting to note,
however, that all of them include the virgule.
The punctuation features listed are symbols. That is, they have no
independent meaning of their own. Any meaning that they have to a reader is
purely a learned convention. Obviously, these symbols are both used and
misused. Every aspect of a writing system can be used as an element of style
whether it involves a spelling, a word, word order, a punctuation feature,
sentence length, or anything else. In this study we are concerned with the
more general use of the punctuation features and how they are used in a
majority of writing situations rather than some specific genre. Why should
we want to establish a new paradigm for punctuation when some others
already exist? Thomas S. Kuhn has also provided an answer for that. He said
it is “to urge a change in the perception and evaluation of familiar data”
(Kuhn, 1970:viii-ix).
23
1.8 Summary
Punctuation, then, is viewed as controlled by some variation of speech
qualities or in some way by some grammar of English. Attempts to define it,
if there are any, are usually very general or couched in some metaphoric way.
None of these are explained in detail. Instead, most works that include
information on punctuation devote their attention to listing rules that must
be followed in very specific locations where punctuation might be needed.
24
Chapter 2
The Evolution of Punctuation
2.1 Introduction
In this chapter we will trace the development of punctuation over the
centuries since the beginning of writing. This should provide a sense of the
early purpose for the use of punctuation and how that has changed over time.
There have been different reasons for its use that match the different uses for
writing. What is unusual is that some of the reasons for its early use have not
disappeared as the purpose for writing has changed over time.
The punctuation system we have today was not designed by a group of
linguists or designers but instead gradually evolved over several millennia at
the hands of craftsmen. It was designed, in a way, by the writers themselves,
the scribes and later technicians involved in the printing trade, who
gradually saw some ways to make reading more efficient and understandable
by adding symbols to the language they were making visible. Looking at the
way punctuation evolved may offer some insight into what they were seeking
to accomplish by adding symbols so that we may better understand its uses
today.
25
2.2 The Earliest Punctuation
According to Bruce G. Trigger, the earliest forms of writing had little to do
with speech. Instead, they were mostly bureaucratic records that were
necessary to meet the needs of the growing complexity of civilization. There
was no need to record anyone’s real or imagined speech and, consequently, no
need to show how it might be organized to represent how someone might say
it (Trigger, 2004:44). As such, we might think there was no need for
punctuation.
It is, of course, difficult to establish exactly when punctuation first
began. Many languages which used it may have disappeared and left no
trace. However, Rochelle Altman, in her Absent Voices: The Story of Writing
Systems in the West, has looked at languages in the middle east, a place
where many written languages had their origins, and decided that the
Sumerian Cuneiform could very well contain the oldest form of punctuation.
The Sumerians figured out a way to represent words and incorporate several
sets of features with them:
By 2450 [BCE], cuneiform writing limits and punctuation were
stabilized. Each word was encased in its own box (case) and
arranged from right to left. Inside each case, words were spaced
as spoken or “uttered.” The right hand margin (incised line)
served as a stop.
(Altman, 2004:23.)
26
The line at the right at the right edge of the box, which serves as a stop, is in
effect the first punctuation, she claims.
Altman also notes the use of what may be punctuation in other earlier
written languages. The colon turns up in Phoenician, for example, and it
continued to appear in other written languages over the centuries (Altman,
2004:38). Apparently, there is something mystical about the colon’s
appearance that attracts scribes, and it is still around today. The Etruscans,
she claims, adopted spacing and paragraphing by the eighth century BCE
(Altman, 2004:39). In this period, the door seemed to be open for
experimentation in the way to organize and display writing.
However, it isn’t until later, when writing systems attempt to make
records that represent speech, that punctuation becomes even more
important. The alphabetical approach that culminated in the Greek system
(c. 750-480 BCE, according to Daniels and Bright, 1996:271) could be seen as
a marvelous achievement in communication, but alphabetical language
systems also pose problems for discussions of punctuation, we shall see,
perhaps because they may represent the spoken language too well.
2.3 Writing as Technology
One tendency is to see writing as little more than an exact recording of
speech instead of the translation of language from one technology to another
technology in a different medium. Walter Ong says:
27
. . . we find it difficult to consider writing to be a technology as
we commonly assume printing and the computer to be. Yet
writing (and especially alphabetic writing) is a technology,
calling for tools and other equipment: styli or brushes or pens,
carefully prepared surfaces such as paper, animal skins, strips
of wood, as well as inks or paints, and much more.
(Ong, 1982:81-82.)
Those of us who can read and write often forget that there are places in the
world where scribes are very necessary craftsmen. M. T. Clanchy, in From
Memory to Written Record: England from 1066 to 1307, 2nd ed. (1993: 114144), goes into great detail in describing the skills and material needs of a
person who had to write a thousand years ago. Just because we can use a
keyboard and printer today doesn’t make the task that much easier, however.
We still need to know how and why to punctuate.
2.4 Problems with Terminology
Perhaps the most difficult problem in studying about punctuation is the lack
of clear definitions of features of the language. This begins, as we have noted,
with “language” and “writing” and stretches on to “paragraph,” “sentence,”
and even “word” at times. (Prefixes have at times been separated from words
by a space, for example, and might not have been counted as words.) We
might be told that punctuation is used to aid elocution. That may mean that
28
a text is punctuated so that it reveals how it should be spoken out loud.
Terms like “prosodic” and “intonational”—which are synonymous in many
reference works—are used to describe the way a text is structured,
apparently for oral delivery. In other places the same terms are described as
synonymous with “suprasegmental,” which refers usually to the stress, pitch,
and juncture in a sentence.
When the term “rhetorical” is used to describe punctuation, it is often
used without any explanation. If it is defined, it is usually done in such a way
as to be equally misunderstood as referring to phrases and clauses. In the
more traditional sense, rhetoric usually describes larger entities, mostly the
units in the way an argument is organized, not the order or character of
words or phrases in a sentence. Rhetoric is traditionally one of the members
of the trivium which consists of the studies of grammar, rhetoric, and logic.
Apparently, these terms were being used as categories for determining the
appropriate home for punctuation because all of them turn up in descriptions
of punctuation in one age or another.
Naomi Baron, however, in “Commas and Canaries,” claims that there
are two types of punctuation: 1) rhetorical punctuation, which refers to the
attempt to create an oral rendition of a text; and 2) called grammatical,
syntactic, or logical, which look at the structure of the text instead (Baron,
2001:21-22.)
29
Vivian Salmon has offered a definition of “rhetorical” as well:
. . . ‘rhetorical’ punctuation [is] designed to show where and for
how long to pause for breath, or how to show emphasis and tone
of voice.
(Salmon, 1962:358.)
Scholes and Willis say it in a slightly different way:
In the elocutionary function, punctuation serves as a set of
instructions for reading a text aloud . . . a kind of
phonetic transcription for prosody (stress, pause, and
intonation). In its syntactic function, punctuation serves, like
other aspects of orthography, to convey meaning . . . by
identifying lexical elements and clausal, phrasal, and sentential
structure.
(Scholes and Willis, 1990:13.)
The elocutionary function, after the beginning of silent reading, apparently
transformed into the rhetorical version, which is still around.
Vivian Salmon has also offered a way to classify punctuation:
Fundamentally, all punctuation is a method (albeit a very crude
one) of conveying meaning which is not expressed lexically; this
meaning may be of three kinds: (1) grammatical, indicated by
punctuation in its ‘separating’ function, whereby units within
the sentence are marked off from one another and sentence
distinguished from sentence; grammatical punctuation also
places the sentence within one of the categories of statement,
30
question, or exclamation/command; (2) emotional, marking a
speaker’s attitude to a statement; (3) logical, indicated by
punctuation in its ‘linking’ function, by which is shown
the degree of closeness in the semantic relationship between
structurally independent grammatical units.
(Salmon, 1979:47-48.)
To this, Salmon adds an interesting comment about another way of
classifying punctuation:
. . . a very valuable and somewhat different system is
postulated by H. Whitehall, who sees that the purpose is ‘to
make grammar graphic.’
(Salmon, 1979:48: Whitehall, 1956:119)
Finally, “sentence” is just as puzzling. In one approach, it simply can
be defined as “the meaning,” and a meaning can take any one of a number of
forms. An even simpler definition comes from Thomas Tuite, author of The
Oxford Spelling book: Being a Complete Introduction to English Orthography:
“A sentence is words put together as they ought to be” (1726:116). M. B.
Parkes also reminds us, when we are reading older manuscripts, that words
like “sensus and sententia are medieval ones which are not necessarily
equivalent to the modern English ‘sense’ and ‘sentence’. Moreover, the terms
could even mean different things at different times during the Middle Ages”
(Parkes, 1978: 131).
31
If it is difficult to find good sources for information on punctuation in
ancient and medieval times, Ian Robinson, who is working on medieval
punctuation, has gone further and publically solicited help in finding
information about a century which included the incunable period in English:
“ . . . and if anybody knows of a history of punctuation, especially as regards
the English language between about 1450 and 1550, I would be most grateful
to be informed” (Robinson, 1992:36). The incunable period lasted for just a
few decades after the start of printing in England and could be interesting to
a scholar because it might reveal the struggles of printers to make sense out
of the transition to punctuation in the new medium.
Robinson is particularly interested in the concept of the sentence, the
available definitions of which he finds unsatisfactory. “Periodus is a word
well established in both Latin and its Greek original, but again it is hard to
be quite sure what it means syntactically.” After considerable research, he
said: “What I have not found before the Renaissance is any clear statement of
the modern concept of the sentence” (Robinson, 1992:41). He concluded that
the sentence seems to have developed in the seventeenth century:
For most of a thousand years English prose, and Latin written
in England, was punctuated by rhetorical stops, not
grammatical marks. They altered remarkably little between
Wulfstan and Caxton. The scribes evidently thought of the units
of prose not as we do, syntactically, but as phrases, constructed
32
by the voice. The whole thing is voice- and rhythm-based, not
syntax-based.
(Robinson, 1992:45.)
Robinson is one of the few writers to state this quite explicitly.
The literature on the sentence is vast. But Joan Persily Levinson tried
to sort it out somewhat by labeling a visible sentence, the words between a
capital that is apparently the beginning and some terminal punctuation
ending it, as an “orthographic sentence” in the attempt to establish clearly at
least one name for a sentence in a definite visible form. She also tried to
avoid the term “fragment,” which has traditionally had bad connotations, and
replaced it with “partial,” a more polite word for a sentence that seems to be
lacking all of its parts (Levinson, 1985:126-127).
Levinson also traced the idea of the “pause” to Aelius Donatus in the
fourth century CE, who supposedly recommended three kinds. The pattern
got repeated for centuries in discussions of punctuation and ultimately
became connected to the comma, the colon, and the period, even though their
meanings now are not intended to be related to sound (Levinson, 1985:26).
You can probably still find this three-part description of pauses as
representing punctuation in a handbook on a bookstore shelf today. In an
attempt to quantify one of the smaller gaps you might make in your delivery
or reading, the term “half-pause” was also invented (Levinson, 1985:28).
Punctuation features can even be classified strangely. Some writers
mistakenly label punctuation features as “graphemes.” Graphemes, however,
33
are usually defined as graphic versions of phonemes, which are a
recognizable class of sounds. Punctuation features have no individual
phonemic character themselves, though they might modify a grapheme. (See
Rogers brief discussion of this, 2005:15.) However, the term “grapheme” has
survived the arguments of some linguists who see no need for it.
2.5 The Genre Confusion
In addition to problems with terminology, punctuation is frequently, not
seldom, discussed as if all kinds of writing are the same, regardless of the
historic period. It isn’t. The difference in the writing style between a poem, a
play, a treatise, and a parts catalog can be extraordinary . Yet discussions of
punctuation often proceed as if the writing in them is all the same and the
audience is all the same so the punctuation is all the same.
Percy Simpson, in his Shakespearian Punctuation, took modern editors
to task because of their careless editing of Elizabethan poetry and drama—
and rightly so (Simpson,1911:7). But even mixing poetry and drama is
dangerous. The voices in poetry are usually those of the poets. The voices in a
play are multiple, and the language and punctuation have to be adjusted to
reveal that. (Having written scripts for a telecourse, I am aware of the
differences that have to be taken into account for the various participants
who have to read what is on a teleprompter. All of the voices are
considerately more like speech than like writing so the dialogue and
34
punctuation will appear to be more chaotic.) Research in which the
punctuation of many genres are mixed together may produce curious
conclusions. Some scholars have been careful to avoid this problem, however.
Editorial preferences are also a concern. One cannot tell if a writer’s behavior
comes from personal preference or a company style manual.
It is apparent that any discussion we might have about punctuation
will be labored by the lack of clear definitions and understandings about
what punctuation is scientifically, how the terms used in describing it are
defined, and what context it is in. So the challenge is to put punctuation in
some place, some definite category, and describe what makes it function
there. Doing that may provide some benefits related to the analysis of writing
in many different contexts and for many different reasons.
2.6 Characteristics of the Evolution
Once punctuation was established in some form, there were a number of
events that altered it in some way. One was some sort of change in the
language or an improvement in the readability based on the appearance of a
written page. Another was the change in the size of the audience of readers,
mostly because of more instruction. There was also the interest of some
significant person, someone who had the ability to promote or improve
punctuation, which then had an influence on its development. The most
significant changes, of course, occurred because of the sudden availability of
35
printed materials which presented what were standardized approaches to
punctuation at the time. This obviously promoted feedback, which
undoubtedly had some influence on all of the people involved in the writing
and the production of printed materials. At that moment, the needs of the
reading audience began to outweigh—or at least prove to be an equal match
to—the philosophy and habits of the writers, editors, and printers.
2.7 Greek Punctuation
Earlier, I avoided the term “paragraph” as a part of the discussion of
punctuation, focusing instead on what occurs mainly in a sentence (in the
modern sense of it). But there could be an argument in favor of including the
paragraph as punctuation in some periods in history because of how it was
being indicated. For example, the early Greek writing used a number of
methods for showing paragraphing involving the use of size or space, one of
which is still used today: allowing space at the very beginning of a paragraph.
Obviously, space was costly in earlier centuries. So they also sometimes
projected a letter into the margin, left an extra space mid-line, or enlarged a
letter to show the beginning of a new paragraph. (Thompson, 1966:60.) A
symbol for a paragraph was also used in Greek writing as well (¶), and it was
usually located marginally. This is more like an indicator of a section of a text
which has no linguistic influence otherwise today.
36
Almost immediately upon locating a study on punctuation, one is liable
to encounter yet another terminology problem. For example, E. M.
Thompson’s discussion of Greek paleography puts aside such symbols found
in early manuscripts as “Dots or points, single, double, or treble, [which are]
seen in ancient inscriptions [which are] marking off several words; but these
are marks of separation rather than of punctuation, unless, perhaps, we are
to except those which happen to stand out at the conclusions of sentences”
(Thompson, 1966:60). One of the symbols he is talking about looks
suspiciously like our modern colon and seems to be closing a sentence but
also is used in other capacities, he says. Thompson also acknowledges the
unsystematic use of symbols in Greek generally. But we are left on our own to
discern the difference between symbols that separate and those that
punctuate. In modern discussions of punctuation, separation is often cited as
one of the reasons for it.
But Thompson claims that a “more regular system was developed in
the schools of Alexandria,” probably by Aristophanes of Byzantium in around
260 BCE:
This was the use of the full point with certain values in certain
positions . . . : high point . . . , equivalent to a full stop; the point
on the line . . . , a shorter pause, equivalent to our semicolon;
and the point in the middle position . . . , an ordinary pause,
37
equivalent to our comma. But this system does not appear in
practice in extant papyri.
(Thompson, 1966:60.)
Over time, it is claimed, the middle point disappeared and was replaced by
the comma. The difficulty of finding suitable examples of punctuation in older
versions of languages is an unsolvable problem, of course.
The three places in which punctuation was used in ancient Greek
might be familiar to a modern reader. The symbols seem as if they are
framing the complete sentence as we usually know it, marking its beginning
and end and perhaps some kind of phrasal activity internally in the sentence.
Although the construction of the sentence itself is still being debated by some
scholars, this seems to be an early acknowledgement of a unit that occurs
even in Greek quite frequently. When we have a model for the way
punctuation works that could be applied to the ancient Greek writings, we
might be able to discover what their purpose was for the middle point. The
others seem evident.
Aristophanes was interested in other linguistic features of Greek as
well as syntax. He invented some diacritics—acute, grave, and circumflex
accents—“to assist students, particularly foreigners, in correct pronunciation
of Greek words” (Diringer and Olson, 1989:1046). Early on, it seems,
educators were interested in improving their languages and the learning of
them. Certainly, silent reading was not yet what written works were used for
with these ancient writings. Diringer and Olson note that it is commonly
38
believed that the Roman alphabet was derived from the Greek, but they
think that “the Etruscan alphabet was the link between the Greek and the
Latin” (Diringer and Olson, 1989:1047).
To understand the state of punctuation for about two thousand years,
from about 100 BCE to about 1900 CE, a scholar might need to rely on only
two books, E. Otha Wingo’s Latin Punctuation in the Classical Age (1972),
and Pause and Effect: Punctuation in the West (1993), by M. B. Parkes.
Wingo covers the glory and decline of the Roman Empire, and Parkes picks it
up from there and carries it to the nineteenth century. Both books provide
many examples of manuscripts. [Unfortunately, neither book has an index.]
2.8 Punctuation and the Roman Empire
The tumultuous events of the past have made the search for any more
ancient documents containing punctuation close to impossible, but there are
other reasons why we may never find a manuscript originally written by its
author. According to Parkes:
No manuscript containing a work in the author’s own
handwriting has survived from Antiquity; this absence of
autograph material has been attributed to the practice of
dictating one’s works, letters, and even one’s notes, to
amanuenses.
(Parkes, 1993:10.)
39
The style of the writing is sometimes even unusual by modern standards.
Some authors sound as if they are directly addressing the potential readers,
and even though the punctuation and marginal comments might be provided
by the authors, it was done as readers rather than as writers (Parkes, 1993:
10).
-
Roman writing originally used centrally located dots to separate words.
“But during the third century BCE,” according to Altman, “the Romans
adopted the Greek model and abandoned their previous practices. They
changed from point as word divider and bar as sense divider [in· this·
manner] to a true scriptura continua” [inthismanner] (Altman, 2004:40).
Most of us today would see this as a regression, but an argument could be
made for it to be at least a tradeoff. The disadvantage of points is the visual
clutter. The disadvantage of scriptura continua is the difficulty of locating the
words. But scriptura continua actually has several other advantages in a
culture that believes that writing is merely speech written down. [For the
sake of consistency, the various forms have been regularized to scriptura
continua in this document.] So what would scriptura continua sound like if it
were read out loud? In the way it is heard, speech is like what musicians call
legato, a continuous sound. Modern writing with spaces is more like staccato.
If a person spoke to us in a staccato fashion, we would be very puzzled, but to
“record” speech in a continuous or legato fashion might seem quite normal.
Another advantage to using scriptura continua is that it is more economical;
40
you can get more letters on a page. Until paper became available, the
surfaces for writing on were always hard to find and expensive to prepare.
The use of scriptura continua did not prevent the Greeks from including
symbols around the letters as instructions to readers, so punctuation was still
used but in a contracted way.
So, where others might see disadvantages in scriputura continua,
Parkes has recognized its advantages:
The merit of scriptura continua was that it presented the reader
with a neutral text. To introduce graded pauses while reading
involved an interpretation of the text, an activity requiring
literary judgement and therefore one properly reserved to the
reader. In ancient Rome, readers of literary texts were mostly a
social elite, whereas full-time scribes were usually freedmen or
slaves. . . .
(Parkes, 1993:11.)
Since the manuscripts were intended to be read out loud to people later, it
was proper to get ready for it by inserting your own punctuation. Here less
emphasis is on what the text might mean and more on the way the reader
might care to deliver it, an important aspect of public speaking. “It isn’t what
you say but how you say it” is still being taught in speech classes.
41
2.9 The Glory of Roman Punctuation
Wingo prepares us for his discussion of the punctuation of earlier Roman
manuscripts by offering a definition:
The term ‘punctuation,’ in the restricted sense in which I shall
use it here, refers to the use in writing of certain signs to show
the end of a sentence or to indicate its structure or the
interrelation of its parts for the sake of clarity to facilitate
reading. Marks may also be inserted, again for the sake of
clarity, to show pauses in speech even when no
syntactical consideration would demand them.
(Wingo, 1972:14.)
Like others who have done research on punctuation, Wingo then complicates
matters a bit by including and excluding features without providing a more
explicit definition; in this case he includes the paragraph. But he leaves out
the “interpunctum as a word divider,” the dot or space between words (also
known as the “medial point”), because it was “taken for granted and
universally used during the period in which we are interested” (Wingo, 1972:
14). Indeed, its presence is actually the test for the accuracy of classifying the
documents for the period of his interest. The Romans retained the dots from
the language they inherited. It wasn’t until later that they abandoned them
in about the second century CE and adopted the scriptura continua .
42
Wingo’s argument against using the interpunctum (the medial dots)
as punctuation is that they can’t be both punctuation and something to only
separate the words. When they no longer separate words, then he could
consider them punctuation. He also excludes decorative features and
“lectionary signs,” which are never clearly defined. He includes spaces as long
as they are “divisions within paragraphs” (Wingo, 1972:17). At one point,
Wingo actually says that “The earliest form of punctuation for sense appears
to have been a blank space” (20), but the concept of “sense” is never fully
explained either. (Later on we see it used as “meaning.”) Although we cannot
get a good impression of all of these features discussed without seeing them,
it is apparent that the creators of the manuscripts were making efforts to
deal with a number of the linguistic forms which they were punctuating to
clarify what the authors meant.
The manuscripts that Wingo is working from may not represent
everything that was written at the time, of course. What exists is what later
people deemed worth keeping:
It is a reasonable inference, therefore, that those writers
preserved by quotation only passages which they deemed
relevant to the ‘modern needs’ of a decaying civilization, and
therefore discarded as otiose references to a system of
punctuation that had become obsolete long before their day.
(Wingo, 1972:22.)
43
Wingo was unable to locate any definitive descriptions of the system as well,
but there are extant manuscripts later in the course of the empire that seem
to “approximate the rules given by the grammarians,” but the grammarians
themselves are from a later time, he says (25-28).
A single text by Augustus (63 BCE -- 14 CE), of which there exists a
number of examples, was used by Wingo as a test for determining the goal of
the punctuation. In the text, he found seven different symbols, only two of
which we might recognize today: spaces and the virgule (32-33). He
determined that the various features were used for four different reasons:
1. to make the document easier to read;
2. to show the end of complete thoughts—probably sentences;
3. to show syntactical relations and set off clauses; and
4. a variety of symbols being used to represent what appeared
to be the same kind of activity.
(Wingo, 1972:48-49.)
Some of these features seemed to function like modern colons and semicolons.
Full stops seem to involve three different symbols. Another seems to be used
to punctuate a series, and other symbols indicate different syntactic activities
within a sentence. In many cases, of course, the exact goal of the symbol is
difficult to determine (Wingo, 1972:49).
Wingo looked at a variety of other documents in order to confirm some
of what he found in the Augustan document and found little that was
conclusive about the use of punctuation in what was available. He did find
44
that legal documents tended to use longer sentences and that, though Latin
documents contained a variety of features, the most common symbol was the
virgule, which was used to show the ends of paragraphs, as a full stop, a
comma, semicolon, a colon, dash, and as a rhetorical pause for emphasis.
(Wingo, 1972:94-131).
His conclusions about Latin writing in the classical age through the
first century CE involved a variety of factors that emerged from the
modifications that were being made to the language in Latin writing:
1. separation by word, sentence, and clause was frequent;
2. the more literary a document, the more punctuation;
3. punctuation in documents by well-known authors may have
shown elements of style;
4. word separation and punctuation “for sense” were related;
5. twenty-one different symbols (including space) were used
within paragraphs;
6. seven was the largest number of symbols found in one
document, but usually it was between two and four;
7. although the symbols weren’t interchangeable, there was
great overlapping in the function for which they were used;
8. despite the overlapping, a tendency to indicate three common
features: a break in a paragraph, a sentence ending, and a
comma;
45
9. only a few texts seem to show a modern-like consistency; and
10. the available materials available for study are not reliable.
(Wingo, 1972:132-133.)
The available evidence, then, suggests a period of trial in the attempt to use
punctuation in some meaningful ways to help readers discern the intended
meaning of a document, not necessarily inserted by its author, but by others
who may have been intending to read it out loud. There appears to be no
clear standard in the way individual symbols were used, except that
identifying the paragraph, sentence, and the clarification of some internal
structures was considered an important thing to do. Counting the number
and type of punctuation features in documents seems be one way of
investigating their use, a practice that is still going on. Punctuation was even
used to make clear the items in a series, however, a fact which could reveal
something about our own approach today.
Wingo noted that documents using scriptura continua were already
appearing by 66 CE, and by 200 CE the interpunctum was probably no longer
the norm (Wingo, 1972:83).
2.10 The Empire after the Fall
One of the most certain things about the history of punctuation in the Roman
Empire in the early centuries CE is that the facts are hard to determine.
46
According to M. B. Parkes, in his book Pause and Effect: An Introduction to
Punctuation in the West:
The attitudes of grammarians and rhetoricians to the
punctuation of texts have been embodied in theoretical
discussions down the centuries, but these treatises have to be
employed with great caution as sources of information about the
usage of punctuation. The principles advocated in the
discussions often do not correspond with the practices
manifested in the bulk of surviving manuscripts and
printed books.
(Parkes, 1993:4.)
His argument sounds as if it could even apply today. The judgments provided
by commentators on punctuation often suggest they have either a limited
view of the kinds of documents they are relating to, perhaps because of a
bias, or they are just being unclear about the documents and the views they
offer. Many just assume that punctuation is connected to speech. The old
oratorical history of punctuation casts a long shadow on what people think
punctuation is for, even in the twenty-first century.
To offer some evidence for his own remarks, Parkes has devoted half of
his 327 page book to reproductions of manuscripts. Fortunately, spaces were
returned to writing in Europe in the seventh century—by Irish and AngloSaxon scribes (Saenger, 1997:12). But lest we think spriptura continua has
become obsolete, Saenger reminds us that the Vai ethnic group in Liberia
47
uses a version of it today with no “word separation, diacriticals, punctuation,
or the presence of initial capital forms” (Saenger, 1997:4).
Standardization of punctuation has always been problematic, and
nobody was in charge in the waning days of the Roman empire. “ Two scribes
can copy the same text and place punctuation in the same positions, but
employ different symbols, or apparently apply different values to the same
symbol” (Parkes, 1993:2). It would take some large steps by people with some
power over the activities of the scribes and users of their work to produce
some changes in the way punctuation was used as well as some useful
innovations in text practices to begin improving the way punctuation was
used. One long-lasting innovation was the minuscule.
2.11 A Small Change with a Big Effect
When a technology is working in some acceptable way, there is usually no
attempt to completely replace it, but users are always looking for way to
improve it. The usual way is by some kind of modification. According to The
Blackwell Encyclopedia of Writing Systems, minuscule writing was not a new
invention, just a modification:
Minuscules were for the most part derived from the earlier
MAJUSCULES, especially from the uncial style, a book hand.
They first appeared in Roman manuscripts of the third century
CE. Their popularity gradually increased, and in the mature
48
form of the Caroline minuscule of the eighth century . . . they
became the major form of European writing.
(Coulmas, 1999:338.)
Few of us perhaps notice that the lower case b is just a truncated version of
the capital B, that the h is just missing a piece; or that the d is a shrunken
and turned around version of the capital D, that the y is just a shrunken and
moved down version of Y. Like any technology, writing follows the rule of
parsimony and tries to use space as efficiently as possible for multiple
meanings without distorting them too much.
Sometimes an improvement is created in a backward fashion, like the
unanticipated consequences of an action. The beginning of the use of the
minuscule script may have had the effect of creating a new punctuation
feature. Once all the other letters in a sentence were now minuscule (like
lower case), the scribe now could make the first letter of a sentence a
majuscule (like a capital), thus creating a new way to indicate a sentence
beginning and eliminating the need for a symbol underneath a letter or an
empty space to announce the beginning of a sentence. (Chapter 5 will include
some more simplification ideas that have been proposed.)
2.12 The Early Middle Ages
Throughout the period after the fall of the Roman empire, oral reading
continued to be the norm, and scriptura continua continued to be the format
49
for a page of writing. Only minor changes took place in the appearance of
texts, such as the introduction by Cassiodorus in the sixth century of “red
marginal notations designed to help the reader find discussions of important
topics” and dividing texts into chapters, where spaces were sometimes used
in the chapter headings but not in the rest of a text (Saenger, 1982:376).
In 1972, Niles Eldredge and Stephen Jay Gould went to a conference of
paleontologists and presented a new idea involving evolution. They intended
to point out that there were some glitches in Darwin’s approach to speciation.
They had observed situations where gradualism was not always the case.
They proceeded carefully, knowing that attempts to alter existing theories
might bring an outcry (Eldredge and Gould, 1972:81-115).
They were aware of the ideas of Thomas S. Kuhn, who warned of the
dangers of upsetting old theories with new ones, despite their apparent value
(Kuhn, 1970:76-77). According to Kuhn:
As in manufacture so in science—retooling is an extravagance to
be reserved for the occasion that demands it. . . . The decision
to reject one paradigm is always simultaneously the decision to
accept another, and the judgment leading to that decision
involves the comparison of both paradigms with nature and with
each other.
(Kuhn, 1970:76-77.)
Eldredge and Gould did have their idea criticized, but it managed to survive
the criticism. It is now known as the “punctuated equilibrium,” the idea that
50
sometimes events occur which cause something to take a leap in its
development, something “macroevolutionary,” whether it is genetic or
technological. As Gould puts it, ‘‘the story becomes more like ordinary history
in the crucial sense that predicable components, driven by the internal logic
of a system, interact with peculiar contingencies to yield a result that none
could have anticipated” (Gould, 2007:303).
Apparently, just such a set of peculiar contingencies came together in
the British Isles:
The origins of rapid, silent reading lie in the scribal techniques
and grammatical teachings that developed in Ireland and
England in the seventh and eighth centuries. The first separated
Latin manuscript books in western Europe were Irish . . . . In
eighth-century England, we find grammatical treatises
composed in separated script by Anglo-Saxon authors trained by
Irish masters that begin to offer direct insight into the
pedagogical implications of word separation.
(Saenger, 1997:83-84.)
Throughout first few centuries of the new millennium, English was
already in a gradual transformation from Anglo-Saxon to Middle English.
After the invasion of 1066, the language of the powerful was French, which
was considered culturally superior to the English of the time (Kuteva, 1999:
224). But as French words were absorbed into English, the royalty took a
51
bilingual approach, and by 1400, English was allowed in the courts and
Parliament (Kuteva, 1999:224). In that century, English faced yet another
major change, but more suddenly.
Perhaps three more punctuated equilibria that have something to do
with writing, but only indirectly, occurred after the beginning of silent
reading: some changes in the appearance of letters, the invention of printing,
and reading online today. In the millennium after word separation occurred,
manuscripts remained places for art work and highly decorated letters. Even
after printing began, some complex fonts made even printed materials hard
to read. In the twentieth century, finally, some attempts were made to
eliminate fonts that were overelaborated and devise some more quickly
discernable fonts. Fractur was eliminated from German printings, for
example, and Helvetica, a rather streamlined and now widely used font, was
invented. Studies were also done on the differences between serif and sans
serif type faces.
Printing created a large audience that was likely more sensitive to the
way writing was punctuated and probably made it known, affecting how it
was then improved. As Walter Ong put it, “Print gave to visualist
organization of thought and to textuality . . . a force unknown before, and in
doing so effectively served pedagogical expediency and at the same time
dissociated knowledge from discourse and gave it a quasimonologic setting
(Ong, 1983:vii).
52
Locating an idea buried in a text has always been a problem. The first
help came from France: “In the late tenth and early eleventh centuries in
Reims, Fleury, and Chartes, résumé notes evolved into marginal treelike
schematic diagrams,” the precursor to our table of contents (Saenger, 1997:
79). Today, one can use a “search” mechanism—if it is available—to find a
word anywhere in a digital text located somewhere in the world. Meanwhile,
“page” has become a purely spatial term for a quantity of text that fills a
screen, and “links” may soon completely replace “footnotes.”
2.13 The Role of Important Individuals in the Middle Ages
Oral delivery was the interest of Quintilian, a first century CE rhetorician,
who defined the comma, colon, and period for his day. The period expressed a
complete thought, “but the orator must be able to deliver it in a single breath”
(Baron: 2001:21-22).
Another person who had an influence on the development of
punctuation was Magnus Aurelius Cassiodorus in the early sixth century,
who created a program of study for scribes. The use of punctuation was still
for pauses, but it was used to indicate meanings of certain words to help
locate the meaning of a text. “According to Cassiodorus punctuation is ‘clear
pausing in well regulated delivery’ . . . [as well as] a guide to the
interpretation of a text” (Parkes, 1993:17).
53
Isadore of Seville (c.560 – 636) apparently saw writing as a technology
that spoke directly to the mind through silent reading rather than being
spoken out loud, and he arranged the order of the comma, colon, and period,
which had been used in a different way to that time (according to rhetorical
principles) for silent reading (Parkes, 1993:21).
In the late eighth century, Alcuin became an assistant to
Charlemagne, whose idea was to promote scholarship. Alcuin tried to revive a
kind of punctuation which had fallen into disuse involving distinctiones and
subdistinctiones. The symbols involved pauses, but they were placed at
different heights in a text based on importance, meaning, and the
incompleteness of the “sense” in the body of a sentence (Parkes, 1993:303304).
After printing started, the need for some kind of punctuation standard
was imperative. The solution came from Aldus Manutius, an Italian printer,
who chose five symbols for the comma, semicolon, colon, period, and question,
intending them to be used in a grammatical system of punctuation. The
popularity of this set of symbols spread and was used by many printers.
Unfortunately, not all printers used them that way, which once more
contributed to the confusion in texts (Baron, 2001:42).
54
2.14 The Ninth Century Turnaround
The transition away from speech as the model in the use of punctuation thus
occurred before the end of the first millennium:
By the ninth century readers and scribes had come to perceive
the written medium as an autonomous manifestation of
language, which was apprehended as much by the eye as by the
ear. Since punctuation had ceased to be solely a form of direction
for the oral performance of the written text, more emphasis was
placed instead on the identification of marks with pauses in the
process of reading silently as well as aloud: such marks enabled
readers to recognize the constituent grammatical structures of a
Latin text, and to evaluate the roles which these structures
played in communicating the message of that text.
(Parkes, 1993:34.)
In an article on the problem a modern editor of a medieval writing faces in
representing punctuation, Mary-Jo Arn summarized it this way:
Because medieval punctuation was unstandardized, there is no
simple correlation between specific medieval and modern marks
of punctuation. Because such marks may, at times, serve to
indicate breath pauses, poetic caesuras, the insertions of
numbers into text, or warnings to later copyists, they sometimes
cannot even be called punctuation in our sense. And of course
55
different genres of medieval texts enjoyed different punctuation
traditions—and some no punctuation at all. Medieval
punctuation is therefore a most unreliable guide to modern
editorial punctuation of medieval texts.
(Arn, 1994:162.)
John Lennard found the same problem in his examination of punctuation in
the twelfth century:
By the twelfth century there was a wide variety of punctuation
in use, which is best to regard as overlapping repertoires of
punctuation, each repertoire being associated with a particular
scriptorium or a particular geographical area. These variant
repertoires were gradually absorbed into a general repertoire, of
which the principal marks were the punctus, punctus elevatus,
punctus interragativus, and virgula , approximately
corresponding to the modern full-stop, colon, question mark, and
comma. These correspondences are of function rather than
shape, for the shapes of the marks did not become standardized
until the economics of printing led to the emergence of
specialized type foundries supplying many printers; and the
gradual change of shape still continues. (Lennard, 1991:3.)
The varieties of older punctuation, we can assume then, have a minimal
influence on our contemporary punctuation, so it would not be productive to
56
pursue them point by point. More important was the reason why punctuation
was being used at all. But the use of space as punctuation continued to have
an important influence. The introduction of space between words clearly did
something that went beyond merely an improved ease of reading.
According to Saenger, some practices in the seventh century made for
better reading, not only because of spaces, but also because scribes were
creating lines of text only ten to fifteen letters in length, thus improving the
ability to see and retain a stretch of text better. Modern research on saccades
has found that to be about the optimum length for good reading (Saenger,
1982:378). Consciousness of the value of space apparently increased over the
centuries. According to Parkes, twelfth century scribes not only left spaces
between words but began to reduce the spaces between letters, and they
began to use serifs to link letters in words, which probably had the effect of
making a word appear more quickly as an independent unit. In addition,
punctuation features began to take a more modern appearance (Parkes, 1993:
41).
Once books were put in libraries and chained down in the thirteenth
century, silent reading became the norm, a practice still encouraged in
libraries (Saenger, 1982: 397). Humanist scribes of the late middle ages
invented quotation marks, parentheses, and enhanced capitalization, and
their sentences included the comma and period in syntactical patterns
(Saenger, 1982:410). Parentheses (known as lunulae) were first found in a
57
manuscript from 1399, and their grammatical place in a sentence and what
kind of effect they create have been debated ever since (Lennard,1991:5).
Scribes, it seems, were finding more symbols with which to modify the ideas
in a text.
2.15 The Great Division
As writing and the teaching of it became more common, factions developed.
One group saw writing more as a silent “art” while others saw it more as
representing speech. It was in the thirteenth and fourteenth centuries that
the seeds for the disagreement about the role of writing may have been
planted. What is writing, then? Is writing simply meant to record speech in
some fashion? Or is it meant to transmit ideas from one mind to another in a
tangible form? The scholastics of the period were leaning in the direction of
grammar and logic as guides to understanding the meaning of texts.
Meanwhile, another group had a different idea:
The proponents of ars dictaminis in the thirteenth and
fourteenth centuries reacted strongly against the scholastic
attitudes to language. They were concerned with the whole art
of prose composition: to stimulate interest in a correct and
elegant Latin, and to put forward rules for the writing of letters
in a highly formal and ornamental style. They invoked the
oratorical ideal with its aural response to the written word. . . .
58
Reactions to scholastic attitudes to language were
accompanied by reactions against the punctuation found in
scholastic texts . . . .
(Parkes, 1993:44-45.)
Thus, the war between speaking and writing as the preferred model for
punctuation may have begun six centuries ago and is still raging.
2.16 The Humanists of the Fourteenth Century
The last major influence on the role of punctuation before the advent of
printing involved the humanists:
The humanists were the successors of the dictaminists. Both
were equally concerned to encourage the writing of correct and
elegant Latin; both favored the epistle as a vehicle of
communication.
(Parkes, 1993:46.)
Latin, which had been the language of scholarship for at least 1500 years
already, was of course the proving ground for experimentation in
punctuation. The vernaculars could imitate the Latin approach to
punctuation of the humanists or find their own way. The humanists revered
some works of Cicero. Nostalgia set in, and the humanists tried to emulate
his style:
His letters provided them with new models, new ideas, and new
ambitions. In their own letters they began to write of their
experiences and opinions in a subjective manner in the first
59
person. By imitating the works of ancient classical authors, and
by appropriating their attitudes, the humanists sought to
recreate ancient ideals.
(Parkes, 1993:46-47.)
The resulting confusion led to the development of new marks, which began to
appear around 1400 CE. The semicolon may have made its debut around that
time as a compromise: it was half colon and half period; perhaps something
between a comma and a complete stop.
2.17 The Modes of Medieval Punctuation
Parkes’ assessment of punctuation of the middle ages is that it took two
different directions: diectic qualities or equiparative qualities. Diectic was
likely to produce different interpretations of a text in different copies, while
equiparative was likely to produce more uniform texts. For example, in two
texts of the deictic type, “the punctuation has modified the ‘dynamics’ of the
same text to achieve different, specific interpretations, although inevitably
there are overlaps between them.” But in the equiparative texts, the
“extensive pointing can indicate grammatical or rhetorical structures in more
detail than diectic punctuation [and] the frequency of the punctuation marks
makes it easier for a reader to construe the text . . .” (Parkes, 1993:71). A
third way, a ”neutral reading of a text,” was achieved by “indicating only the
most basic divisions of the text—paragraphs and sententiae—leaving the rest
unmarked” (Parkes, 1993:71-72). Most important to the scribes, he believes,
60
was “to elucidate sense” and involve grammar or rhetoric only if necessary to
achieve a certain interpretation (Parkes, 1993:72).
Arguments about whether or not punctuation was an accurate
reflection of speech in the writing in Latin was not an issue in the middle
ages because there were really no longer any native speakers of it. But the
ideas of the humanists about punctuation continued to hold sway even during
the early days of printing.
The most important difference that occurred when printing began was
that all the copies of a text in a print run had the same punctuation. The
focus could now be on the punctuation itself and the reason for doing it, not
the individual behavior of an author, scribe, or orator. The earliest printers,
according to Parkes, used the punctuation they found in the manuscripts, but
some standardization began to occur in the fifteenth century already. Having
arrived late, the semicolon was not treated with as much courtesy as other
symbols, but it became a favorite some time later. “By the 1580s there was
clear evidence that compositors were responsible for introducing punctuation
marks—especially the semi-colon—to replace others indicated in an author’s
copy” (Parkes, 1993:53). The technicians had begun to exercise their role as
modifiers of the technology.
61
2.18 The Impact of Printing on Punctuation
Like anyone who tries to spread a new idea or a new technology, there is
always the thought of how the users of the old ideas and technology might
respond. Why William Caxton in 1476 set up his press in Westminster
instead of London has been a topic for discussion ever since. Some have
speculated that he did not choose London because “the stationers and
professional copyists who were already entrenched there would not welcome
a printer, whose ‘black art’ might threaten their livelihood” (Childs, 1976:
150). A more likely reason was probably that he thought Westminster would
be a better neighborhood for selling books (Childs, 1976:152).
Consider the problems he faced: setting up machinery, finding people
to set type and run the machine, selecting manuscripts and editing them to
prepare them for print, and still having to worry about things like
punctuation. According to Saenger and Heinlen, “The late incunable printer,
by establishing a close and rigorous control over punctuation and nuances of
meaning that flow from punctuation, became . . . an editor in the modern
sense of the word” (Saenger and Heinlen, 1991:255). Readers apparently
began to accept the editing role of the printers and did less to repunctuate
printed books:
Throughout the Middle Ages, readers, even long after a book
had been confected, felt free to clarify its meaning through the
addition and modification of prosodiae, punctuation, and
62
marginalia. Under the influence of printing, reading became
increasingly an activity of the passive reception of a text that
was inherently clear and unambiguous.
(Saenger and Heinlen, 1991:253-254.)
In effect, there were two versions of punctuation in operation at the
time, one for Latin, which was taught in the schools, and one for vernacular
English, which was not taught in the schools and was less systematic; add to
this the variety of kinds of English Caxton faced: The Old English Bible,
Beowulf, Chaucer, and Wyclif’s Bible (Bolton, 1982:178):
So there were three kinds of punctuation represented among the
four texts. Both Old English manuscripts were almost
unpunctuated; the few marks of punctuation were so infrequent
and sporadic that their absence would make no difference to the
understanding of the text.
(Bolton, 1982:178)
According to W. F. Bolton:
Caxton was the inheritor of over a thousand years tradition
about punctuation, most of it based on the needs of breathing
but with increasing attention to elocution and syntax. Within
less than a hundred years after his death in 1491, punctuation
had altered to a system that is even now clear to modern
readers.
(Bolton, 1982:180.)
63
Thus, another punctuated equilibrium had taken place. But Bolton was
perhaps too optimistic. In the next centuries, two different theories for the
use of punctuation developed. Some standardization did occur, however, and
some new features were added.
An Italian printer gets a lot of the credit for the changes that took
place, including the arrival of parentheses:
Lunulae were included in the fonts designed by the great
humanist printers, notably Aldus Manutius and Pietro Bembo,
and were disseminated in print, reaching England in 1494.
(Lennard, 1991:5.)
A work in Latin containing the parentheses, Opus Grammaticum by
Sulpitius defined “five types of punctuation—full-stop, question-mark,
virgula, colon, and lunulae—and provides examples of each mark . . . “
(Lennard, 1991:5). This was probably not lost on the printers, and things
began to change in the punctuation of English printed work.
2.19 Syntax vs. Rhetoric: The Theoretical Argument
The standardization of punctuation got a boost in the mid-1500s from two
publications by John Hart in 1561 and 1569 which provided detailed
recommendations for using punctuation, but his descriptions mixed rhetorical
and syntactic descriptions, assigning different lengths of time to pauses for
different features in musical terms (Salmon, 1999: 22).
64
At about the same time, the son of Aldus Manutius published a
document on punctuation in 1561 that got great interest, De ratione
interpungendo. (The work can also be found with a slightly different name,
Interpungendi Ratio, and in translation, in Punctuation: Its Principles and
Practices by T. F. Husband and M. F. A. Husband. They identify it as part of
a long treatise on orthography. Their version is seven pages long. The
examples in their translation seem reasonable, but the punctuation in any
translation could be problematic. See Chapter 3.)
According to Mindele Treip:
The period between about 1580—1680 . . . witnessed a
particularly marked evolution away from rhythmical and
oratorical, or sometimes theatrically dramatic, concepts of
syntactical design, concepts which encouraged individuality and
flexibility of expression, toward more logically and
grammatically oriented views. The approach was
increasingly to the ideal of ‘correctness’ in writing: toward
regulating all aspects of composition toward universal and
logically fixed standards. Altered practices in punctuation are
among the features which reflect these changes.
(Treip, 1970:x-xi.)
Over the centuries since Caxton, a good number of documents which
include advice on punctuation were published, many of them significant
65
grammars that were popular in their own time. Between 1500 and 1800,
many arguments took place over certain punctuation features and their
placement in a sentence. Authors lined up on one side or the other of the
rhetoric-or-syntactic argument or didn’t care to decide between them.
Scholars looking at different periods after 1500 tend to come up with
different views of the rhetoric-or-syntactic argument. Walter Ong surveyed
seven publications printed between 1582 and 1692 in search of the policy
they might support:
Mulcaster (1582), Puttenham (1589), Heywood (1612), Gill
(1621), Butler (1634), Daines (1640), Jonson (1692)
Virtually all of them mentioned breathing or rhetoric in reference to
punctuation. According to Ong :
the most telling characteristic, then, of Elizabethan and
Jacobean punctuation theory remains the fact that it never cut
itself loose from the traditional view of punctuation as basically
a physiological rather than either an elocutionary or a
syntactical (logical) device.
(Ong, 1944:360.)
Here and there one person stands out. Arguments about the number of
punctuation features and where they should go seemed to involve all the
commentators. Sometimes curious distinctions in punctuation features were
made. For example, John Hart subdivided those points having a correlation
to spoken language into two types:
66
(a) those representing pauses in speech
(b) those representing intonation: those were the marks of
interrogation, exclamation, and parenthesis
(Salmon, 1988:298.)
Salmon believes that the rhetorical view dominated the eighteenth century
(Salmon, 1988:295). She surveyed a number of grammars of the time to find
out if it were true:
Hart (1551), Morley (1597), Hume (1617) , Butler (1636), Daines
(1640), Lewis (1672), Luris (1672), Clare (1690), Aickin (1698),
Brightland (1711), Kirby (1711), Mattiare (1712), Watts (1721),
Kames (1762), Sheridan (1762), Lowth (1762), Steele (1775),
Cockin (1775), Steel (1786), Bayly (1789), Murray (1795),
Cockin (1775), Walker (1785) Fogg (1796)
(Salmon, 1988:passim.)
Salmon determined that perhaps most of them, except for Brightland, who
was somewhat uncertain, ascribed to the rhetoric approach. Attempts to
classify features according to some pattern of use also occurred in these
books—and are still occurring, judging from any survey of writing handbooks.
In the later decades of the eighteenth century, the war of the points
apparently began, and attitudes toward punctuation changed more quickly:
During the period c.1789-1824 the grammarians evolved no
consensus about punctuation, but their exchanges sharpened
67
the quality of analysis. By mid-century many significant
typographical conventions . . . had been established in that
particular combination which still widely endures.
(Lennard, 1991:144.)
Greta D. Little disagreed to some extent with the findings of Salmon,
who apparently saw a slow but continuous drift toward rhetorical
punctuation. Little also saw a drift, but beginning in the 1600s, toward
grammar, which culminated by 1850 (Little, 1984:371). Little claims not two
but three traditions were developing: “rhetorical, grammatical, and
typographical” (Little, 1984:372). The rhetorical was concerned with sound
and emphasis and had no hard and fast rules; the grammatical, which has to
do with sentence structure, has fairly rigid rules; the typographical has to do
with “the history and development of printing and the division of labor
between author and printer in matters of punctuation” (Little, 1984:372). She
believed that the works of twentieth century scholars contributed little new
information to help us understand punctuation, but listed some exceptions (a
few of which will be reviewed in Chapter 3).
2.20 Punctuation in Concert with Music
For two thousand years prior to 1700, writers and had already been looking
for some way to characterize the use of punctuation. When it was related to
elocution, there was little room to question its need. But when documents
68
were gradually used less and less for oral delivery, the justification for
punctuation became a troublesome question because someone had to make a
decision about why and where to insert symbols into a text. Reading out loud
could still be used as a practical reason for using punctuation, but silent
reading changed things. Now a reader needed punctuation to help
understand a text. Meaning rather than delivery had suddenly become more
important, but, for some unknown reason, readers were not able to recognize
that they were “looking” at language and not “hearing” it. The result was the
search for abstract principles that might explain and help sort out and justify
the use of punctuation. Every abstraction in the trivium, for example—
grammar, rhetoric, logic—was mined for justifications for punctuation, but
not one of them could fully explain its purpose. Readers and scholars kept
trying to find ways to make them work, however.
After printing, the choices between two features of the trivium,
grammar and rhetoric, were the final contenders. But around 1700, another
paradigm was proposed: music. According to Stephanie D. Vial, in her
dissertation, Take Pause: Musical Punctuation in the Eighteenth Century, the
analogy of music and language seemed evident. Both used syntax and
phrasing; longer units in music resemble some in written form. The sonata
has a beginning, middle, and end, just as an essay does. Perhaps music, with
its semidemiquavers and Gestalt-like forms, could be used to explain
punctuation and return it to an earlier day as a help in both oral delivery and
69
reading. (Leonard B. Meyer, in Emotion and Meaning in Music, discussed the
applications of Gestalt to music and found it wanting, however (Meyer, 1958:
85-86).)
Serious attempts were made between 1700 and 1850 to ally music and
the other parts of the trivium, logic and rhetoric, to writing, but before the
end of the nineteenth century, all of them were trumped by grammar.
2.21 The Triumph of Technology
Three authors who have tried to make some sense of the feud between the
advocates of the rhetorical and the syntactic approaches to punctuation and
the changes they may or may not have brought about are Park Honan (1960),
David Cram (1989), and Karsten Schou (2007). In reviewing the history of the
feud, each came to some kind of conclusion. Honan believes that the syntactic
theorists won, partly because of books published by the Wilson company.
Cram is not sure that a battle even existed: “The supposed conflict between
the rhetorical and syntactic principles is either spurious, or has at least been
blown up out of all proportion” (Cram, 1989:310). Schou took a common
position of many writers between 1600 and 1900: “The general experience is
that syntax has been central at least since 1600, although prosody played and
still plays a certain part” (Schou, 2007:213).
In spite of Cram’s view, publications that were beginning to emerge
during the nineteenth century that had something to do with punctuation
70
tended to make the argument clearer and imply that the grammaticists had
won the argument. It was not so much what the proponents of the
grammatical approach were saying as much as the way they were saying it
and who was saying it that seems to reveal something.
The Wilson printing company might have become interested in
publishing their own book on punctuation as a result of their printings of the
grammars of Lindley Murray, whose works became quite popular around
1800. The first edition came out in 1795 and others followed. Murray’s
grammars had large sections on punctuation, and the books themselves were
over four-hundred pages in length. Murray was careful about how he treated
punctuation, apparently aware of the arguments about what kind of
principles controlled it. In the third edition of his grammar of 1816, he was
careful to avoid conflict by adopting a compromise:
As punctuation is intended to aid both the sense, and the
pronunciation of a sentence, it could not have been exclusively
discussed under the part of Syntax, or of Prosody. The nature of
the subject, its extent and importance, and the grammatical
knowledge which it presupposes, have induced us to make it a
distinct and subsequent article.
(Murray, 1816:389.)
(Strangely, a printer’s guide published in the United States in 1818 by C. S.
VanWinkle quotes Murray’s careful approach but then talks about pauses.)
71
Perhaps the Wilson company saw an opportunity to profit from an
interest in educational books at the time. Their work titled A Treatise on
Grammatical Punctuation, by John Wilson, was first available in short form
in 1826 and went through at least thirty-two editions by 1899 under the
name A Treatise on English Punctuation. Wilson’s introductory remarks
reveal that he was not afraid to take on the proponents of the rhetorical
approach and others:
We would not overrate the importance of Punctuation, or deny
that many subjects are worthy of a higher regard, and have a
more immediate and vital influence on the well-being of society.
. . . It is intimately connected with the principles of grammar;
subservient to the purposes of syntax; essential to the clearing
up of ambiguities, which so often obscure composition; and
useful to the more ready understanding even of those sentences
whose construction is not liable to the charge of obscurity.
(Wilson, 1864:3.)
The Industrial Revolution was likely an exciting time for proponents of
punctuation as well as people manufacturing machines. A book by S.
Rousseau in 1813 also promoted the grammatical approach:
The grammatical construction of a language, which ought to be
the basis of Punctuation, has seldom been considered as
adequate to the purpose: too much accommodation to the reader,
72
and too little attention to the Rules of Grammar, have usually
been the sources whence the doctrine concerning the points has
been deduced. But such principles, with defence to those
Gentlemen who adopt them, we can conceive to be erroneous
however specious; as indeed must be all systems that are
founded upon varying authority. . . . Punctuation should lead to
the sense; the sense will guide to modulation and emphasis.
When Punctuation performs its office thus, it will point out
likewise the grammatical construction of a sentence; for the
sense of a passage and its grammatical construction are
inseparable.
(Rousseau, 1813:31-34.)
Rousseau’s interest in promoting the concept of punctuation seems obvious.
The book begins with a twenty-nine page history of punctuation with
references to books he had studied at the British Museum.
A century later, the grammatical approach is still strong. Printer
Theodore Low DeVinne, in his 1902 book on punctuation, is not afraid to
debunk some of the old practices while he is promoting the proper views:
The function of points is to make expression intelligible.
Punctuation tries to do this by separating the words that are not
closely related, and by keeping together those that are related.
Incidentally points justify rules of grammar, but the
demonstration of those rules is not their first purpose. Points
73
have small elocutionary value. The old teachings that there
should be one pause in the voice after a comma, two pauses after
a semicolon, three pauses after a colon, and four pauses after a
period, are now generally condemned.
(DeVinne, 1902:245.)
The tenor of the time was probably that expressed by Wilson in his
1844 edition, and Wilson probably deserves the last word on the issue of
punctuation in his time:
Notwithstanding, however its utility, Punctuation has not
received that attention which its importance demands.
Considered merely as the plaything of the pedant, or as the
peculiar function of the printer, it is often neglected or perverted
by those who have occasion to present to the eye either their
own thoughts or the thoughts of others.
(Wilson, 1844:4-5.)
Wilson proceeds to take on the grammarians, the lawyers (who garble the law
by leaving out the punctuation), the painters, engravers, and lithographers,
the letter-writers, as well as the people the printers publish:
Even the author—who, of all writers, ought to be the most
accurate—not infrequently puts his manuscript into the
printer’s hands, either destitute of grammatical points, or so
74
badly punctuated as to create a needless loss of time to the
compositor.
(Wilson, 1864:5.)
We can wonder if any of the handbook writers of the twentieth century
happened to notice what Rousseau, Wilson, and DeVinne had to say.
2.22 Conclusion
Thus, we arrive at the twentieth century with a relatively stable set of
symbols for punctuating written English yet a difference of opinion about how
those symbols perform in a text. But people were at least beginning to take
an interest in looking into what the underlying principles were in their use.
In chapter 3 we will examine some of the linguistic studies of punctuation in
the period that was labeled The Century of Progress to see if there was any
breakthrough in the views on the purpose for punctuation and how it
accomplished that purpose.
75
Chapter 3
Punctuation Scholarship in the Twentieth Century
3.1 Introduction
When you are looking to understand an entity, you usually want answers to
some basic questions. How is it defined? What does it consist of? What is its
scope? What makes it work? The existing scholarship on the entity also
shapes your search for understanding. You can accept what already is
believed and go on from there, or you can locate a new paradigm. At the close
of the nineteenth century, there were basically the two approaches to
punctuation that were available and being debated on and off: the rhetorical
approach and the grammatical approach. Most studies before 1900 had
accepted one or the other and did some variation on the use of punctuation
that did not really answer or deal with any of the basic questions.
It is not unusual to encounter the complaint in some book or article
that not enough is being done in a certain field of study, but to find that
comment bridging two different centuries speaks to the neglect that has
occurred in that field. It turns up 150 years apart:
Punctuation has not received that attention which its
importance demands.
(Wilson, 1864:4.)
. . . punctuation has much to contribute to language processing
by both humans and computers. However, perhaps in part
76
because of these difficulties, there has been surprisingly little
research in the area.
(Doran, 1998:1.)
This kind of comment occurs in other studies of punctuation as well, and it
says something about the lack of serious interest in the principles behind its
use. A lot of it had to do with the state of linguistic studies during the early
part of the century, but some may also have had to do with the belief that one
of the views concerning punctuation had already been accepted more widely
and that there was no longer any need to do any serious research.
3.2 A Century of Little Progress
Writing itself was not being treated very well by noted linguists in the
twentieth century, and books that were used for teaching punctuation seem
to have simply accepted one or the other of the major theories in describing
how to use punctuation without entering the debate. The surprising thing
about the twentieth century, then, was that it did not add much to our
knowledge of how punctuation works. First of all, there are very few books
that could be called scholarly studies of punctuation. There are only two in
the twentieth century that use the word “linguistic” in their titles. Even those
books seem reluctant to get completely immersed in all of the possible
answers to the way punctuation might be linguistic by limiting the scope of
their research in some way. Perhaps the best two books about punctuation
published in the century came out in 1905 and 1990. In between, there are
77
few published works that add to our understanding of punctuation as a
linguistic feature. The 1990 book, by Nunberg, did not spawn more books on
punctuation immediately, but it did stimulate some interesting research that
ended up in dissertations instead.
There are, of course, many “how-to” books on punctuation and many
that are simply entertaining. There are a number of scholarly books on just
one limited area of punctuation, perhaps one feature, the punctuation of one
author, or how to teach it, some of which will be mentioned briefly below. On
the other hand, there were a number of dissertations on various aspects of
punctuation that did not get published in book form, but most of them seem
to have tried to avoid the major debate and deal with uncontroversial
features. Otherwise, the vast majority of books on punctuation that were
published were books that did not discuss the linguistic aspects of
punctuation although they might preface their lessons with some brief
discussion of the issues.
3.3 Published Works on Punctuation in the 1900s and early 2000
3.3.1 Punctuation: Its Principles and Practices, by T. F. Husband and
M. F. A. Husband, published in 1905, is really like no other book on
punctuation. It begins with a single sentence written originally by Thomas
De Quincy in his Confessions of an English Opium-Eater but shown in
scriptura continua form. It is a “cumulative” sentence of eighty-six words held
78
tightly together in majuscules with no punctuation and is considerably hard
to read. (De Quincy’s original version, which is in an appendix, used one
capital, one exclamation, three dashes, one semicolon, four commas, and one
point for an abbreviation.) The first few lines look like this:
HOWPAINFULTOSEEORTOKNOWT
HATVASTREVELATIONSOFGRANDE
URANDBEAUTYAREWASTINGTHEM
SELVESFOREVER . . .
(Husband and Husband, 1905:2.)
The point that the Husbands were trying to make about the value of
punctuation got made very quickly.
The first fifty-three pages of their book are a history of punctuation
ending at about 1900. Recognizing the dominance of Latin in the history of
English and its development, they chart the problems along the way that
involve punctuation. Not until Alfred the Great’s efforts to get English
translations made do things begin to improve. The difference between Latin,
an inflectional language, and English, a word order language, causes
problems. English, it turns out, probably needs more punctuation, but there
is little uniformity in its use.
A major shift in the use of punctuation occurs after the start of
printing. But Caxton, perhaps the best-known of the early printers in
England, seems to have had no system for punctuation and used only the
79
comma, semicolon, and period. The picture changed when a document by
Aldus Manutius, an Italian printer, circulated in Europe. It set forth a
coherent system that got adopted widely.
To fill in the gap to the present, the Husbands include twelve brief
summaries of books on punctuation from 1617 to 1859 and provide each
writer’s approach to punctuation theory. The on-going debate between the
rhetorical and grammatical theories is mentioned, of course. They note that
the sentence as a form is frequently a part of discussions of punctuation,
perhaps implying its importance in the future for punctuation changes.
Their final summary is only partly hopeful:
Grammatical classifications and rules tended towards
absolutism. This, we take it, is one reason for the revolt from
strict constructional punctuation that is noticeable in the latter
part of the nineteenth century. But the chief reason is to be
found in the fact that a return to more native forms of
expression has made heavy constructional punctuation
unnecessary.
(Husband and Husband, 1905:52.)
They devote a chapter each to the full stop; exclamation and question;
comma; semi-colon and colon; and the dash, parenthesis, and quotation,
using examples from various writers to show how each feature is used.
80
The real bonus to the book, however, is the inclusion of a translation of
the Interpungendi Ratio by Aldus Manutius, the document that may have
spurred the standardization of punctuation among printers. They claim their
translation is close to the original punctuation. (However, there are at least
two ways to translate a document: by word or by meaning. Either way, it may
not be able reveal anything about the way the punctuation is used because it
is likely to be different from one language to another.) The document does
contain one each of a comma, semicolon, colon, capital, and question symbol.
Their approach to punctuation is that it is an “aid to intelligibility”
and:
The difference between good pointing and bad is, in such a
theory, mainly a question of degree, rightness being determined
by adherence to the precise degree of minuteness with which we
decide to break up a sentence . . . .
(Husband and Husband, 1905:127.)
The Husbands did provide a service by including the Interpungendi Ratio by
Aldus Manutius as an appendix because it is otherwise difficult to find. It is
surprising that the book was never reprinted, because it provides a brief but
thorough background to the current punctuation climate at the time, but it is
presently available as a free EBook through Google.
81
3.3.2 Modern Punctuation: Its Utilities and Conventions, by George
Summey, Jr., is lengthy and puzzling. He seems at first to take a very liberal
approach to punctuation. In the Preface, we are told that “The so-called rules
of punctuation, as a general code for all conditions, have not worked
(Summey, 1919:vii). This is supported later when he says “The current rules,
moreover, are too numerous and too rigid” (Summey, 1919:2). Later we are
told that “There is no single working principle. In cases of punctuation it is
necessary to apply one or more of several considerations. There are questions
of custom, clearness, emphasis, movement, economy, variety—sometimes
even of appearance on the page” (Summey, 1919:4).
Even the way of identifying the symbols comes into question:
The use of the term to mean a punctuation mark is practically
obsolete; and punctuation in the rare sense of observing stops
with the voice is aside from the present purpose. Punctuation
marks are meant for the eye. Though they may convey
suggestions of intonation and vocal pauses, that is not their
usual purpose.
(Summey, 1919:19.)
By the time we get to page six, we find out that “Punctuation is not a matter
of mechanical correctness: it is an art” (Summey, 1919:6). With that, we
expect that Summey will be about to adopt an aesthetic approach to
punctuation, but he doesn’t.
82
Instead, the “rhetorical nature of the marks must be insisted on,
because the grammatical viewpoint—legitimate in itself—has laid emphasis
upon formal syntax rather than upon communication” (Summey,1919:31). We
are told that “The inaccurate and misleading classification of punctuation
marks into grammatical and rhetorical points, or into grammatical and
grammatical-rhetorical, is still current” (Summey, 1919:24). He also rues the
separation of punctuation from the rest of writing in textbooks and instead
sees it as a feature that is used in paragraphs, and he devotes a few chapters
later to a discussion of paragraphs and the way sentences are involved in
them.
Summey then gets immersed in the grammatical vs. rhetorical
argument and, during a lengthy examination of just about all the
punctuation features one by one and offering evidence from a variety of
writers, he finds most of them to be clearly rhetorical. He also mentions a
number of earlier works on punctuation and finds agreements and
disagreements with them in how punctuation functions. He also offers
another definition: “Punctuation marks are signs which indicate the
relationship and character of the words which they precede, enclose, or
terminate” (Summey, 1919:20). But we are not sure how this applies to his
discussions of the individual punctuation features. The definition is difficult
to apply, except that we should probably look for some kind of rhetorical
83
consideration in a situation and use it for punctuation. A later definition of it
also leaves us confused:
Rhetorical means aiding or defeating in whatever degree any of
the aesthetic or practical aims of writing. More specifically, it
means aiding or hindering communication in respect to
clearness, economy, ease, agreeableness, force, persuasiveness,
or whatever may be desired. A rhetorical use of a point is simply
an instrumental use; instinctive perhaps, but in its degree
effective, whether for or against any of the purposes of writing.
(Summey, 1919:26.)
The goal of the final chapter is “to point out some facts about the
frequency of punctuation marks in representative current books and
periodicals” (Summey, 1919:241). The punctuation features used by a variety
of popular writers are counted and compared, apparently to see how the use
of certain features influences their styles, but no conclusions are drawn from
the data.
Summey deserves some praise for three comments he made along the
way: “Punctuation marks are made for the eye” (Summey, 1919:19). “White
space judiciously employed makes communication easier and more
pleasurable” (but he assigned it no role) (Summey, 1919:26). “Often the only
way to avoid an awkward pointing is to revise the phrasing” (Summey, 1919:
84
23). Of course, he is right about revision. Too often, punctuation gets the
blame when a complex construction should have been rewritten instead.
Summey’s first book on punctuation is like a bellwether of some
coming discussions of punctuation theory. It struggles to prove punctuation
belongs to one class but is always mentioning another one, as if there is
continued uncertainty about the accuracy of the classification. The evidence
that seems to lurk in the background of his discussion is that it is difficult to
reject one approach to classification completely and adopt another one.
3.3.3 Summey’s second book, American Punctuation, seems somewhat
bolder. It is “a study of punctuation in its relationship to the art of writing”
(Summey, 1949:v). His opening definition reminds us of a hundred
handbooks, though it includes one wrinkle: “Punctuation is the use of
conventional marks for the purpose of making written matter clear at sight”
(Summey, 1949:3). He classifies the kind of writing he will study as prose,
but not fiction or advertising. He also omits from consideration any feature
that alters the appearance of a grapheme, such as capitals, italic, font
appearance, and the like.
Mostly he is interested in “structural punctuation,” which he identifies
as “paragraph breaks and the various marks used between or within
sentences to make a writer’s meaning clear at sight” (Summey, 1949:v). It
includes the usual features of punctuation that are found in the horizontal or
85
syntactic line of a sentence (. ? ! , ; : -- ( ) [ ] . . .). “Punctuation,” we can recall
now, “is primarily for the reader’s eye, with only partial correspondence to
the movement, stresses, and inflections of speech” (Summey, 1949:6).
Summey tries to cancel out the old ideas, but only partly: “The various marks
have no definite pause values, and even relative values are variable”
(Summey, 1949:7). He claims that the old approach to punctuation as related
to oral reading is unworkable:
All the marks have rhetorical effects, and all of them do their
part in making grammatical relations clear—not for the sake of
grammar but for the sake of meaning. They are grammatical
because they help to make clear the relations between sentence
members, between expressions in series, between a noun and a
following descriptive modifier, etc. At the same time they are
rhetorical because they are useful for clearness and emphasis.
Bad punctuation, of course, is bad rhetoric.” (Summey, 1949:8.)
At this point, we are uncertain about where the lines are drawn. This seems
like a compromise—or a unity. Punctuation, he maintains, should be seen as
part of the writing process, and paragraphing is also a kind of punctuation (a
view that some later writers refer to as “macro-punctuation”). He anticipates
Joan Persily Levinson’s definition of the “orthographic sentence”:
86
A sentence is a group that properly begins with a capital (not
following an interior mark such as a semicolon) and ends with a
period or other terminal mark.
(Summey, 1949:23.)
A large area of the book is devoted to how the “structural” points
operate in a sentence, including all of the syntactic symbols mentioned above,
along with samples of writing and comments on these features by other
writers. Some space is devoted to morphological features like the hyphen,
apostrophe, and the abbreviation dot, in a deviation from his attention to
“structural” points.
Again, we have a final chapter that involves “quantitative” information
about how various writers and various publications use many of the points,
but he admits that it is “too small a sampling of any single writer or
periodical to warrant sweeping conclusions about the general practice of that
person or journal” (Summey, 1949:157).
In both of his books, Summey appears to be unable to forget about the
two opposing approaches to punctuation that competed in the previous five
hundred years and seems to be trying to resolve the argument by isolating
different groups of symbols and allying them with both rhetorical and
grammatical responsibilities in some way. His arguments never quite seem
to come together as a unified theory around a single concept.
87
3.3.4 In 1987, Charles F. Meyer published A Linguistic Study of
American Punctuation, a book version of his dissertation. The study was
limited to American writing and was based on a collection known as the
Brown Corpus, from which he selected twelve samples in three styles totaling
72,000 words. He used the approach of Summey in his 1949 work and
intended to only look at “structural punctuation,” which would consist of
periods, question marks, exclamation points, commas, dashes, semicolons,
colons, and parentheses (Meyer, 1987:xiii).
Features like paragraphing, apostrophes, hyphens, brackets, ellipsis,
quotation marks, underlining, or other incidental uses of punctuation, in
dates, for example, were not included in the study. (By using Summey’s
approach, though, some other features that affect graphemes and morphemes
were also excluded,)
Five earlier studies were investigated to establish a taxonomy for the
study. Those studies produced a list of ideas that punctuation might do:
terminate, introduce, separate, enclose, combine, join, and insert. It was
decided that, broadly, there are two major concepts at work in punctuation:
separating and enclosing. The features that separate are: ? ! : and
sometimes – , (Meyer, 1987:4). The features that enclose are: , -- ( ) (Meyer,
1987:6). Boundaries that the features functioned in are: syntactic, prosodic,
and semantic (Meyer, 1987:8)
88
Meyer decided that an important issue to consider had to do with the
three functions. “In other words, why does punctuation separate and enclose
syntactic, semantic, and prosodic boundaries?” (Meyer, 1987:11.) Meyer
decided that American punctuation was conventional and that the uses of the
features were dictated by the three boundaries identified (Meyer, 1987:15).
The remainder of the book is a chapter each devoted to documenting
the three boundaries and including examples from the Brown Corpus for each
one. Another chapter is devoted to “violations” of the norms and one to the
way style manuals approach the subject of punctuation. Meyer does not come
to any major conclusions about the totality of the subject of punctuation but
lets the examples speak for themselves. He does offer conclusions at the end
of each chapter which provide some wrap-up for each function. For example,
for the syntactic function he offers this:
Punctuation has various syntactic functions. Because the marks
of punctuation are hierarchical, they interact in various ways to
distinguish subordinate from superordinate boundaries.
Moreover, punctuation is a device that serves to segment
syntactically lengthy and complex constituents, particularly
compound sentences and adverbials heading sentences,
subordinate clauses, and the second main clause of a compound
sentence.
(Meyer, 1987:38.)
89
The other chapters have similar conclusions for their function or topic.
Certainly, these conclusions are not that much different from what one might
find in a good manual devoted to punctuation, but the advantage here is that
Meyer provides the evidence immediately to prove the point.
The disappointing aspect of this study, like others, is that once again
only some of the features of punctuation found in a text are being looked at
and assessed for their use. No attempt is made to locate a single, underlying
reason for punctuation in the first place. Part of the problem is that so little
information about the linguistic aspects of punctuation is available. We can
imagine that Meyer did the best he could with what was at hand by
preparing a taxonomy from sources that were both reasonable and available.
3.3.5 Perhaps the most referenced book on punctuation, one which
also appeared close to the end of the twentieth century, was The Linguistics
of Punctuation (1990), by Geoffrey Nunberg. Like other writers on the
subject, Nunberg laments the treatment punctuation has received, even at
the hand of other linguists, and supports the study of it and writing, saying
that it “should no longer be necessary to defend the view that writing is truly
language . . .” (Nunberg, 1990:3). He argues that “punctuation is in fact a
linguistic subsystem, and hence to be considered as part of the wider system
of written language” (Nunberg, 1990:7). Furthermore, “viewed as a whole, the
system has no analogue in the spoken language . . . except for a few
90
overlapping devices,” and should be thought of “as a part of the study of
graphical representation in general” (Nunberg, 1990:9).
Nunberg airs and discounts two views of punctuation that have
prevented more scholarship—the “intonational” and the “transcriptional”
approaches that try to claim that writing is a poor imitation of speech at best
(Nunberg, 1990:11-13). Rather, he says, we should approach punctuation as
“an autonomous system that admits of study in its own terms” (Nunberg,
1990:15). He claims that punctuation should be thought of together with
other graphical features, including font alternation, capitals, indentation,
and spacing, all of which should be labeled as “text-category indications”.
These indications can have three functions: they delimit, separate, and
distinguish (Nunberg, 1990:17).
Nunberg focuses on general kinds of writing below the level of the
paragraph and puts aside special items like block quotes and bulleted lists.
He sets up two grammars to makes his points in the discussion that follows: a
text grammar (which is a form that relates to discourse analysis) and a
lexical grammar (“which is responsible to describe the dependencies that
obtain among lexical items in the text”) (Nunberg, 1990:19). From this point
on, he really deals with only what he calls the “text-sentence” and defines it
as “that unit of written text that is customarily presented as bracketed by a
capital letter and a period . . .” (Nunberg, 1990:22). In other words, it could be
a phrase, a NP + VP sentence, or some variety of compounded sentence(s)
91
(Nunberg, 1990:22). (This is also roughly equivalent to Levinson’s
orthographic sentence.)
A chapter is devoted to a number of “text-category” adjuncts: colonexpansions, dash-interpolations, and parentheticals, paying most attention to
the places where they occur and how they interact with the comma in the
lexical grammar which (we recall) can delimit, separate, or distinguish. Using
these categories, Nunberg explores how the regular punctuation features
(fonts, other type manipulations, and spaces) interact.
Perhaps his most useful (though seemingly obvious) observation is the
sequencing constraint: “no more than one point indicator (i.e., comma, period,
semicolon, dash, or colon) can be presented in sequence in any given position”
(Nunberg, 1990:57). He also speaks of an “absorbtion rule,” under which
features of different types occurring in the same location have a certain order
of precedence” (Nunberg, 1990:57). Parentheticals and quotations obey
special ordering rules as well. This culminates with a series of rules which
dictate the order in which punctuation features are placed (Nunberg, 1990:
70).
The standard symbols, incidentally, can be viewed as affixes or clitics
attached to a neighboring word—the comma being an example of a leftcliticizing mark because “the comma is treated as an inseparable part of the
word to which it is attached” (Nunberg, 1990:58). Generally, Nunberg’s
approach is based on assigning specific rules to each punctuation feature and
92
stating rules for how they all interact. A later chapter provides examples of
how some punctuation functions in certain genres, such as recipes, and takes
a more “discourse” approach to punctuation.
In his conclusion, Nunberg points out that sentences of previous
centuries followed different patterns than those of today and included
punctuation in different ways. He warns that attempts to modernize the
punctuation of older manuscripts are usually misguided and destructive of
historical information (Nunberg, 1990:131).
In general, Nunberg’s work involves the “how” of punctuation more
than the “why.” The different punctuation points operate and interact
according to sets of rules he mostly invented. His approach involves mainly
discourse and the sentence in particular. Nowhere is there a mention of a
phoneme or a morpheme. But his discussion of absorbtion rules stands out as
a valuable concept. It is a concept that never seems to be found in indexes to
help people understand why one feature of punctuation is allowed to swallow
another.
3.3.6 The most recent book involving a study of punctuation is
Punctuation as a Means of Medium-Dependent Presentation Structure in
English (2013) by Sebastian Patt. The first problem we have is determining
what is meant by a “presentation structure.” Jürgen Esser, who was the
inspiration for Patt’s work, put it this way in an article he wrote:
93
Quite generally, presentation structures are the result of
stylistic choices in encoding at the level of form and substance.
(Esser, 2000: 1524.)
The next term we need to define next is “medium-dependent.” A feature can
be medium-dependent or medium independent. Since punctuation functions
only in writing and not speech, it is medium-dependent.
Patt cites a variety of scholars who took the position that punctuation
is essentially rhetorical, thus opening the door to a discussion of punctuation
from that point of view. Punctuation marks, we are told, “can be understood
as ‘indices’ in a basic semiotic sense. They predominantly convey
paralinguistic material” (Patt, 2013:10). (Studies involved with paralanguage
usually put punctuation in the same category as all the nonverbal
communication discussed in relation to speech.) So now every punctuation
event, apparently, “can be described as an interaction of (at least) two
expression systems, with each system having its own distinctive units of
analysis” (Patt, 2013:85.) The two could be the punctuation and the language,
perhaps (both form and content?).
But punctuation also needs definition:
In the present study, the term “punctuation” is synonymous only
to “segmental marks”, i.e. to a set of non-alphanumeric
94
characters that supplement the typographic realization of the
medium-independent word-forms. In principle it is therefore
limited to the orthographic sentence. (Patt, 2013:94.)
The orthographic sentence plays an important role in Patt’s analyses, just as
it has in some others. Punctuation marks “should, at least partly, function as
graphic sign posts, whose choice and placement facilitates the decoding (and
processing) of the structural arrangement of a given text” (Parr, 2013:88).
A major part of the book is devoted to seeing how the features of
punctuation interact with the language in many situations and how they
influence the meaning in those places. The presentation structure is being
diagnosed in these locations, but we need to be careful because “a
punctuation mark does therefore not have a specific, decontextualized
meaning” (Patt, 2013:275).
In his final pages, Patt confesses that “punctuation marks
predominantly convey paralinguistic information” and “the communicative
value of punctuation marks is not (completely) predetermined by fixed
conventions, but within a given framework . . .” (Patt, 2013:275). Patt
believes his approach “moves away from a static rule-based description
towards a more dynamic, context-based interpretation of punctuation” (Patt,
2013:277). To others, however, it might just resemble a somewhat complex
discussion of style.
95
3.4 Books Which Contain Some Linguistic Value. Although there were
few books about punctuation directly linked to linguistics in the twentieth
century, there were still books on somewhat related subjects which still
contain some valuable information for linguistics and punctuation.
3.4.1 Mindele Treip’s Milton’s Punctuation and Changing English
Usage, 1582-1676 (1970). Although the book is mainly about Milton’s work, it
also contains some concentrated information about one period in the
development of punctuation in English.
3.4.2 John Lennard’s But I Digress: The Exploitation of Parentheses in
English Printed Verse (1991). The work includes a lot of information about a
single feature of punctuation over a period of more than 500 The years.
3.4.3 Anthony Graham-White’s Punctuation and Its Dramatic Value
in Shakespearean Drama (1995). The work provides some insight into the
way play scripts were punctuated in the Elizabethan period and beyond. This
was a period when punctuation practices were changing rapidly.
3.4.4 Paul Saenger’s Space Between Words: The Origins of Silent
Reading (1997). The space between words may seem like nothing to some
people, but it was likely one of the punctuated equilibria in the development
96
of reading and a large step in the development of today’s system of
punctuation.
3.5 Dissertations Having a Linguistic Approach to Punctuation
In contrast to the dearth of studies about punctuation and linguistics in book
form, there have been a variety of dissertations that deal with some aspect of
punctuation, some of them inspired by Nunberg’s The Linguistics of
Punctuation (1990). Most are limited to some isolated feature of punctuation
rather than dealing with it in a comprehensive way, or they discuss
punctuation as it relates to some other concept the writer was also interested
in.
3.5.1 David Steegar’s Prosody and Punctuation: A Linguistic and
Experimental Study (1975) could very well have been written several
hundred years ago. It explores seeming parallels between prosody and
suprasegmentals and punctuation but deals with only a few features of
punctuation common to sentences. Although the basic text is in English, it is
actually a study of French writing and punctuation, and virtually every
example in the work is in French, making it impossible to confirm Steegar’s
conclusions by someone who does not know French.
97
3.5.2 Joan Persily Levinson claims she set out to do a wide-reaching
study of punctuation, but in her dissertation, Punctuation and the
Orthographic Sentence: A Linguistic Analysis (1985), she laments the
shortage of good sources for research and the apparent traditional focus on
the sentence (which itself she believes is poorly defined) as the home of
punctuation, to the exclusion of most other places in a text. As a result, she
focused her study on something that looks like a sentence but is “neither
‘complete thought’ nor a ‘grammatical unit’” (107), a sentence in appearance
only which exhibits “informational grouping” (126).
3.5.3 In The typographic contribution to language: towards a model of
typographic genres and their underlying structures (1988), Robert Waller
ultimately makes the case that there is little difference between the
typographic aspects of a text and the punctuation in it. His approach is from
the study of typography, but he comes to the conclusion that both typography
and punctuation share the same function: to provide a context for a text.
3.5.4 Robin Hill’s dissertation, A comma in parsing: a study into the
influence of punctuation (commas) on contextually isolated “garden-path”
sentences (1996), involved an experiment with garden-path sentences to see if
punctuation would really help subjects reading them. The results were
mixed, with some structures providing unclear answers.
98
3.5.5 A dissertation by Bernard Jones, What’s the point? A
(computational) theory of punctuation (1996), was aimed at discovering how
punctuation can help in the processing of natural language, a practice that
had heretofore omitted punctuation. His working definition is “any fact of the
text that is not a lexical item or number [including inter-word space (6)]”
would include punctuation. Jones suggests that, because of the frequency of
punctuation features in written English (one for every fourth word),
punctuation deserves more study. He provides a taxonomy of punctuation
that seems quite usable.
3.5.6 The dissertation by Bilge Say, An information-based approach to
punctuation (1998), involves the study of punctuation but uses primarily just
four features (the dash, colon, semicolon, and parentheses) using natural
language processing (NLP) techniques from corpus linguistics, computational
linguistics, and formal semantic modeling. The study concentrated on the
orthographic sentence and the discourse level. A model that resulted could be
taken as a template for NLP software developers for making use of
punctuation more effectively and revealed some noticeable things about
anaphora resolution and presupposition.
3.5.7 Christine D. Doran’s dissertation, Incorporating punctuation
into the sentence grammar: a lexicalized tree adjoining grammar perspective
99
(1998), takes a strong stand against the idea that punctuation has anything
to do with prosody. Various punctuation situations are discussed, and she
concludes that features she added to the lexicalized tree adjoining grammar
(LTAG) to deal with those situations were workable.
3.5.8 Stephanie D. Vial’s Take pause: musical punctuation in the
eighteenth century (2000) examines how the punctuation of writing was
applied to musical composition. During the eighteenth century, punctuation
was required to meet the demands of both grammar and rhetoric, resulting in
a kind of pointing craze which reached its culmination toward the latter part
of the century, when the analogy to musical phrasing reached its prominence.
3.5.9 A dissertation by Benjamin Grindlay, Missing the point: the
effect of punctuation on reading performance (2002), also focused on the kind
of misreading caused by garden-path sentences. The major goal was to find
out if the use of punctuation, particularly the comma, would provide a
dependable and effective way to help in reading such sentences. The study
revealed that the comma was helpful, but a subject’s reading ability was also
a factor.
3.5.10 The problem of English punctuation (2002), by Karsten Schou,
who is a teacher of translation (Danish to English), was begun when he was
100
unable to find adequate answers to questions he had about punctuation. It
involves a detailed look at the history, development, the available literature,
and the current state of beliefs about punctuation. He concluded that the
cognitive approach to punctuation has gotten little attention, the classical
heritage had been neglected, relevant issues had not been described well, and
rhetoric still “should be granted a prominent position in descriptions of the
punctuation system” (300).
3.5.11 In her dissertation titled Punctuation as symbol: experiencing
archetypal patterns through personal narrative (2003), Gail Emily ArriolaNickell claims that punctuation features “offer abundant possibilities for
psychological and artistic interpretations that transcend semantics, and
these offer a re-mythologizing of punctuation symbols as personal narrative”
(iii). Her explorations of the possible uses of punctuation that are related to
symbolism and psychology may help to explain the way some writers have
used punctuation in a personal and stylistic way, despite the rules in the
handbooks.
3.6 Conclusion
The work of a century of research on punctuation was not very enlightening.
The studies surveyed here did not answer the basic questions about
punctuation very well. The controlling feature was still thought to be either
101
some aspect of the way content was arranged (rhetoric), some aspect of
grammar, or somehow that punctuation might still be tied to speech
(elocution), along with attempts to match nonverbal aspects of speech to
nonverbal aspects of writing. Even a book with “linguistics” in its title offered
no concrete explanation for punctuation but instead treated it as a discourse
feature. It is probably time for a fresh approach to the understanding of
punctuation to find out which one of the two major parts of everything is the
model for punctuation.
102
Chapter 4
The New Paradigm for Punctuation
4.1 Introduction: Content and Form
Everything has at least two parts: content and form. The question that is at
the bottom of arguments about theories is often the problem of determining
which of the two dominates in the way something is constructed, understood,
and used. Punctuation is involved in this problem. When writing was thought
of as merely a note-taking system for oral delivery, there was little argument
about what was dominant, the soon to be spoken words or the images in the
document. The content and its delivery were of great importance.
But times have changed and the written document has matured and
has a life of its own. Yet the argument about which of the two features is
dominant in relation to punctuation has not gone away. Is it content or form?
The alternatives still seem to be sound or sight, prosody or grammar. But
there could be a third way of perceiving what is important to display in the
context of a sentence.
What goes on in speech is described in a number of ways: prosody,
intonation, suprasegmentals, and more. But all of these terms relate to the
way the voice configures the sounds in speech, and that is usually the proof
for those who claim that speech processes are the basis for punctuation. The
forms created by the voice, they say, should be used to insert the punctuation
103
in a written document. The proponents of grammar argue for the
grammatical organization of the words in a written statement as the basis for
punctuation. It is how the parts and pieces of a sentence are put together on
a page that should be considered the basis for inserting the various
punctuation features. What is it that the punctuation is protecting or
preserving? What is the plan?
Both media are using the same language so there should not be any
arguments about the semantic aspects; we should not be arguing about the
content of speech and writing. It is really the form of the medium that is
important and how the language is configured. In every translation, both
content and form undergo significant change. That is true. So we need to look
at how writing presents language and makes it available to a reader. The
main place where we find punctuation is in the sentence and its partials.
4.2 Hypotheses for Form
Here are the hypotheses that are to be dealt with in this chapter:
The basic form of the written English sentence is the canonical (SVO)
sentence.
The graphic form of the sentence in written English is the orthogonal
projection, which has two basic dimensions, the horizontal axis
(dynamic/syntactic) and the vertical axis (static/phonological,
morphological, semantic).
104
Punctuation is a system of symbols used to protect the integrity of the
canonical sentence and make it visible in the graphic medium of
writing.
The comma is used in the dynamic dimension to prevent inappropriate
modification of the sentence.
A number of other symbols are used to modify linguistic forms in the static
dimension.
Using specialized punctuation, interpolations are inserted into the sentence.
Full or partial adjunct sentences are added to a sentence using the
semicolon or colon; the semicolon also functions as an elevated comma.
4.3. The Sentence
People who write are encouraged to use full sentences. But finding a clear
and usable definition of the sentence anywhere can be trying. Part of the
problem may be that the English sentence is not very old, so the punctuation
behavior in one may not have developed completely yet either. Ian Robinson
put it this way: “What I have not found before the Renaissance is any clear
statement of the modern concept of the sentence” (Robinson, 1992:41).
Apparently, anyone interested in grammar and sentences before the
Renaissance would be looking more at Latin than at English. The remnants
of the study of Latin were still around in the early 1900s. Robinson believes
that the “completion of the understanding of our modern sentence is a
105
seventeenth-century development” (Robinson, 1992:43). Prior to that time,
those doing the writing thought of “utterances, or breaths, or discourse. With
medieval prose and verse it is much more useful to think of breaths than of
sentences: begin with the voice, and the written language as instructions to
the voice” (Robinson, 1992:44). The result was that “medieval punctuation in
English was either metrical, or rhetorical, or non-existent” (Robinson, 1992:
44).
The problem with any definition is that there are many ways of coming
at it for different purposes. David Olson aired some of the ways one could
judge the validity of a sentence. It might be “well-formed” if it is
understandable to a listener, or it might be “valid” if it is syntactically correct
(Olson, 1977:259). Some people would find a sentence valid only if it were in
the proper context, regardless of its semantic-syntactic correctness (Olson,
1977:260). To others, it would be a sentence regardless of its semantic or
syntactic content as long as it began with a capital letter and ended with
some terminal punctuation, the kind of sentence examined in some detail by
Joan Persily Robinson, who found in her study that the sentence had “no
commonly accepted definition” even by many of the important names in the
world of linguistics in the early part of the twentieth century
(Levinson, 1985:6):
A central hypothesis developed in this research is that the
“sentence” in writing is not a grammatical unit and is not to be
106
explained by the grammar of the language (accounting for both
the difficulty in defining it and the reluctance to do so), and that
consequently its punctuation does not “mark’ syntax. This is not
to say that there is no grammar, that there are no requirements
on syntactic arrangements and on acceptable writing, but that
the characteristics of the sentence in written English are such
that syntactic rules alone (or even mainly) cannot capture them.
(Levinson, 1985:8.)
Levinson seems to agree with Robinson that the sentence did not really begin
to become important until sometime after the spread of printing around
1500 CE.
But things that begin with a capital and end with some terminal
punctuation have to be called something. Levinson decided to call them
“orthographic sentences” (purely on the basis of their appearance, that is),
but then had to deal with those orthographic sentences that were not
syntactically complete, the ones that are usually marked in classroom
writings as “fragments.” Zellig Harris used the term “residue” (Harris, 1982:
388). The more polite name Levinson gave them is “Sentence-Partials”
(Levinson, 1989:116) The question now is how to decide about the way
punctuation is used in both “orthographic sentences” and “sentence-partials”:
Will it be the same or different?
107
We might expect a 450 page book titled Sentence Comprehension: The
Integration of Habits and Rules by Townsend and Bever to come up with a
clear and useful definition of a sentence that we could use for analyzing
punctuation, but we would be disappointed. After quoting several century-old
offerings by Wundt (1911), they leave us to read the rest of the book to try to
locate the one they might have to offer:
· We cannot define sentences as sequences of words because
there are single-word sentences (e.g., “Stay”).
· We cannot define sentences as word uses that have meaningful
relations because there are meaningful relations within certain
word sequences that, nevertheless, are not sentences (e.g.,
“Monday, Tuesday, Wednesday, Thursday, Friday, Saturday,
Sunday”).
· Hence, the sentence must be defined as a sequence that native
speakers of a language intuitively believe to convey a complete
proposition in linguistically acceptable form.
(Townsend and Bever, 2001:11-12.)
In a later chapter, Townsend and Bever do manage to mention some concepts
helpful to understanding sentences—and to these concepts: the schema and
the canonical sentence.
But some day we may be able to avoid the problem of defining
sentences in writing, according to M. A. K. Halliday:
108
Eventually we shall discard the term ‘sentence’ from the
grammar altogether; it can then be used unambiguously to refer
to a unit of the writing system—that which extends from a
capital letter following a full stop up to the next full stop. . . .
We cannot identify a ‘sentence’ in the spoken language . . . only
by defining it as a clause complex. (Halliday, 1989:66.)
4.4 The Sentence by Parts
Perhaps it would be possible to come to a workable way of understanding a
sentence by seeing how it is constructed. Most taxonomies of it would
probably look something like this:
sentence
clause
phrase
word
letter
Certainly, the visible parts of a sentence are here—even if we use the
orthographic sentence without bothering to define it any more or even if we
added the subject and predicate at the clause level. We can usually find all of
these parts when we see a sentence. But some conceptual approach still
seems to be missing. There is nothing here about what holds all these parts
together, how they are organized, and how they produce a meaning. Some
109
kind of process that determines how the sentence parts become a whole
meaning seems necessary.
Lakoff, in his discussion of The Spatialization of Form Hypothesis,
mentions “linear order” schemas and “up down” schemas but doesn’t go into
much detail about them. However, the written sentence is an image schema
as he describes it, and it is probably a gestalt (Lakoff, 1987:283-284). Paul D.
Deane agreed with Lakoff in how the sentence is likely to be framed:
Specifically, it is claimed that linguistic expressions are
processed as if they were objects with internal structural
configurations. That is, they are processed in terms of certain
basic image schemas, namely part-whole and linkage schemas
critical to the recognition of the configurations which define
complex physical objects. . . . In other words, the Spatialization
of Form Hypothesis treats grammar as a form of imageschematic thought in which words, phrases and sentences are
endowed with an abstract structure grounded in immediate
bodily experience of physical objects.
(Deane, 1991:364.)
Written sentences, then, should apparently be seen and not heard.
110
4.5 The Parsing Process
In the 1960s, a number of psycholinguists began searching for answers to the
way a person processes language, particularly concepts that make it possible
to come up with the right order of things to establish an intended meaning.
In their article The Sausage Machine: A Two-Stage Parsing Model, Frazier
and Fodor proposed a parsing devise that “assigns phrase structure to word
strings in two states,” one which notes six-word strings of words and another
which notes the phrase level; in this fashion a reader acquires the “deep
structure” of a sentence (S −> NP + VP) (Frazier and Fodor, 1978:291). This
would account for the surface level of a text that one can see, but there is
apparently more to the comprehension of a sentence than merely seeing what
is on the page. How does the reader know what order the parts should be in?
Fodor, Bever, and Garrett came up with an answer for that earlier:
An early stage in the perceptual analysis of linguistic material is
the identification of the sentoids of which the input sentence is
composed. By hypothesis, each such sentoid will consist of a
subject-NP and a verb which may or may not have an object.
(Fodor et al., 1974:344.)
They assumed that someone following the surface structure of a text was
unconsciously perceiving the “canonical-sentoid strategy” and assuming that
something ordered as NP V (NP) was probably following the ”subject, verb,
and object of a deep sentoid” (Fodor et al, 1974:345). In other words, there is
111
really a form deeper than the “deep structure” that is guiding the decision to
link the words encountered into a form at the top of the parts list that we
recognize as making up a “sentence.” The list of parts does not account for an
important concept. The sentence is not the most abstract rung in the ladder.
There is one above it or one that should replace “sentence.” With this
information, we might be able to solve everyone’s problem and not call a
string of words a sentence unless it has the subject and verb part of the
canonical form in the right order. Everything else would be a partial.
A number of studies over the years have looked at “garden-path”
sentences, sentences in which the order of words causes a misread, to see if
they would reveal anything about the parsing process. “The garden path
theory, as we have developed it, claims that the parser assigns a single
immediate analysis to an ambiguous fragment or a temporarily or
permanently ambiguous sentence” (Frazier and Clifton, 1996:8). The results
suggested that the readers confused by garden-path sentences were being
fooled into thinking that they were seeing the canonical sentence but were
tricked by the word order. In effect, this showed something very important:
the readers apparently believed that there is a proper order of words in a
sentence and were deceived when it did not occur. The study of disordered
sentences proved that the readers were expecting the canonical sentence and
it did not come.
112
Frazier and Clifton later summarized much of the earlier work, noting
three of the most-studied parsing principles:
Minimal Attachment
Do not postulate any potentially unnecessary nodes.
Late Closure
If grammatically permissible, attach new items into the clause
or phrase currently being processed (i.e., the clause or phrase
postulated most recently).
Minimal Chain Principle
Postulate required chain members at the earliest point
grammatically possible but postulate no potentially unnecessary
chain members.
All of these “follow from the claim that the parser constructs syntactic
representations by applying rules (like phrase structure rules) that create a
single enriched tree structure” (Frazier and Clifton, 1996:8-9).
Other studies have also looked at which concepts are involved in
parsing, and there have been claims that both syntax and semantics are
working together, but they also doubted the existence of “parallel analysis”
and think syntax (or form) probably takes precedence (Frazier and Clifton,
1996:5-7). (Again, one can probably also find some strong arguments for the
idea that semantics comes first in parsing.)
113
There is still another concept that contributes to meaning in parsing.
In syntax it is called “adjacency,” and in visual perception studies it is called
“proximity.” When two items are next to one another in a sentence or nearly
always come in a certain order, the implication is that they belong that way
and have a recognizable meaning together (Moravcsik, 2006;32).
David W. Jacobs calls the gestalt-like idea “grouping,” claiming that it
occurs quickly; that items tend to group by edges; that objects being grouped
do not have to resemble anything “real”; and that the distance between the
objects is important (Jacobs, 1988:17-18). (There may be some support for the
“edge” perception idea for words. An online game circulating now involves
asking people to read words in which the letters in the middle are mixed up
but the letters on the ends are correctly in place, allowing a reader to quickly
recognize a word even though some letters are out of order.)
One thing about parsing is difficult to find information about. Just
what is the process called that seemingly “adds” words together to create a
meaning? What has happened to the words in the sentence? In response to
their grammatical relationship, the words have gone together in some way.
How did they manage to do it? How did we get from a line of words to a
sentence meaning? Terms that are sometimes used to describe the process
are “concatenation,” “accumulation,” “modification,” “accrual,” “incremental,”
“aggregation,” “agglutination,” and others. There is obviously some way in
which the words in a sentence go together to make up a meaning greater
114
than the individual words, but the process cries out for a good term to
describe it.
Walter Sargent Stolz, in his dissertation, Syntactic Constraint in
Spoken and Written English, said that Chomsky (1966) put “models of syntax
(grammars) into three categories: (a) Markov models, (b) phrase structure
grammars, and (c) transformational grammars” (Stolz, 1964:7). Stolz
described the Markov model:
The Markovian model . . . is essentially a finite-state automoton
in which the symbol being produced at any given time is a
function of the previous symbols emitted by the device.
(Stolz, 1964:7.)
There are certainly resemblances between a Markov model and what is going
on in a sentence in English, but it was dismissed by Chomsky as not possible.
However, it remains to be seen if someone can locate a model or a process
that describes the continuous modification that occurs from the start of a
sentence until its resolution as a “complete meaning” at the end.
There is still one issue to resolve related to parsing: Where does a
reader obtain the concept of a canonical sentence that seems to lie invisible
underneath the parsing strategies?
115
4.6 Sentences without Meaning
One way to analyze a problem is to remove some part of it so that another
part may become more visible. Noam Chomsky did that with a sentence:
Colorless green ideas sleep furiously.
(Chomsky, 1966:15.)
Apparently, Chomsky subscribed to Levinson’s orthographic sentence
because he used a capital letter and a period. The initial reaction of most
people to this is that it is nonsense, but most still see it as a “sentence.”
(Some people, though, have looked at it as a variety of figurative language
and tried to defend it as having poetic meaning.) In fact, the sentence has a
curious but distorted meaning. It proves at least one thing: the words are in
the right order. With two positive things, words in the right order and
orthographic features, we have a good reason to call it a sentence. The only
thing missing is meaning—content.
Lewis Carroll’s lines are just about as meaningless but much more like
sentences because they include something that Chomsky left out: function
words.
‘Twas brillig, and the slithy toves
Did gyre and gimble in the wabe;
All mimsy were the borogoves,
And the mome raths outgrabe.
(Carroll, 1999:12.)
116
Carroll provides coordinators, determiners, prepositions, plurals, the copula,
one auxiliary, plus a variety of punctuation, including the semicolon, but it is
all nonsense. Both Chomsky and Carroll seem to prove that you can say
nothing and still have a good sentence. But what they really have done is
demonstrate how a sentence works, including the value of the punctuation.
Chomsky does it by putting meaningful words in the wrong places in relation
to one another (with only “orthographic” punctuation), Carroll by putting
meaningless words in the right places (but well punctuated). In doing so, we
see the form of the sentences in both, even when they don’t mean anything.
In a way, Carroll has actually trumped Chomsky. According to Paul
Saenger:
. . . psychologists have observed that the presence of graphically
distinct short function words, including articles . . . and
prepositions, is very important for organizing eye movement
because such short words are particularly easy to decode in
parafoveal vision.
(Saenger, 1997:45.)
4.7 The Schema
Most people would probably agree that repetition is a major source of
cognition. We watch and we listen and in doing so we accumulate most of
what we need to get along in the world. But how do we store all of the
information we gather, and in what form, so that we can access it when we
117
need it? Scholars in many areas have decided that we manage to store many
kinds of information by using patterns of different kinds, linking concepts in
such a way that they are then immediately recoverable as a group.
Some of these patterns are more complex than others, and they could
involve some activities that we engage in every day and others only on rare
occasions. Also, it is very likely that many or even most of the schemas we
are storing in memory are not even conscious. Often, we encounter a
situation in which we are aware of something “out of order” but are unable to
determine exactly what it is. Some of the names used to describe these
concepts are pattern, frame, script, schema, prototype, stereotype, plan,
exemplar, ideational scaffolding, template, plot, abstract symbolic
representation of knowledge, structure of expectation, a generic concept of
memory, framework, scenario, mental model, story, and we could possibly
include plan and blueprint and some others.
Different areas of study seem to find their own characterization for a
pattern or a name for an idea, based on the probable perceptual context of a
situation, but they have a similar outcome: we store in our memories some
kind of structure that we have experienced and recognize it again when we
encounter it. But the most commonly used word to describe this kind of
structure is “schema” (the plural is usually “schemata,” although “schemas”
has also been suggested). One common experience that is often used as a
118
metaphor for a schema is going to a restaurant; after going to a few, we all
know what the likely order of events will be.
Although some writers on the subject credit Kant and his Critique of
Pure Reason (1787) for the idea, more credit is often given to F. C. Bartlett, in
Remembering: A Study in Experimental and Social Psychology, for promoting
the schema, even though Bartlett himself credits one Sir Henry Head for the
idea—and expresses his dislike for the term “schemata” (Bartlett, 1995:198201):
I think probably the term ‘organized setting’ approximates most
closely and clearly to the notion required. I shall, however,
continue to use the term ‘schema’ when it seems best to do so,
but I will attempt to define its application more narrowly.
(Bartlett, 1995:201.)
Bartlett describes a schema as “an active organisation of past reactions”
which are “operating in any well-adapted organic response” involving “any
order of regularity of behaviour” which have been “serially organized” but
operate “as a unitary mass” (Bartlett, 1995:201). Bartlett’s definition seems
good enough to apply to a large variety of events.
David E. Rumelhart, whose interest is reading comprehension, raised
Bartlett’s idea to a new level. Rumelhart describes schemata as a theory:
. . . basically a theory about knowledge. It is a theory about how
knowledge is represented and about how that representation
119
facilitates the use of the knowledge in particular ways.
According to schema theories, all knowledge is packaged into
units. . . . A schema, then, is a data structure for representing
the generic concepts stored in memory. (Rumelhart, 1980:34.)
Rumelhart listed some of the major features of schemata:
1. Schemata have variables.
2. Schemata can embed, one within another.
3. Schemata represent knowledge at all levels of abstraction.
4. Schemata represent knowledge rather than definitions.
5. Schemata are active processes.
6. Schemata are recognition devices whose processing is aimed
at the evaluation of their goodness of fit to the data being
processed.
(Rumelhart, 1980:40-41.)
So schemata are not stagnant. Rather, they are dynamic, ready to change
when the need arises because our experiences change, and they are
apparently based on real events, not just textbook versions of something.
Another important idea that Rumelhart discusses is “perception.” He is
concerned that we perceive a whole in relation to its parts: “the
interpretation of parts and wholes must proceed jointly” (Rumelhart, 1980:
48). We could easily make a mistake and start to believe that Rumelhart’s
discussion is actually about the canonical sentence and gestalt psychology,
because the things he is saying certainly could easily be applied to them.
120
Lest we get too interested in perceptions, though, Johnson-Laird
reminds us that “a schema is not an image, but a model that underlies the
ability to form an image” (Johnson-Laird, 1983:190). It is not a “perfect”
example of a concept but a “typical” or “default” example (Johnson-Laird,
1980:190).
Also, we are not likely to find some kind of universal schema that we
can apply to every situation, according to Asghar Iran-Nejad, because:
conceptually and essentially, a schema is a domain-specific
relational cluster. Beyond this, however, the concept of schema
remains, theoretically, disturbingly vague. One reason for this is
that the metaphors cognitive scientists use (e.g., "link," “
association," "connection," "pointer," etc.) to refer to the relations
among schema constituents are purely conceptual.
(Iran-Nejad, 1980:3.)
But one more idea that Rumelhart offers could be very useful. He says that
schemata are like parsers:
A parser is a device that, given a sequence of symbols,
determines whether that sequence forms a legal sentence
(according to the rule of some grammar) and, if it does,
determines the constituent structure of the sentence. That is, it
determines which symbols in the sequence correspond to which
constituents of the sentence.
(Rumelhart, 1980:40.)
121
In effect, Rumelhart has provided here something that will work for us in
dealing with punctuation, a tool that will fit the task, an invisible schema
that we can use for looking at sentences and seeing how they are organized in
order to punctuate them. We can use the schema of the canonical sentence to
help us identify the basic constituent construction of a sentence and the
punctuation in it.
4.8 Yaggy and the Canonical Sentence
Using the canonical sentence as a starting point for understanding how to
use punctuation is not a new approach. Elinor Yaggy, in her article titled
“Let’s Take the Guesswork Out of Punctuation,” certainly had it in mind
when she recommended the “basic sentence” as a starting point for learning
how to punctuate:
Before attempting to learn about the biggest bugbears, commas,
the writer needs to learn the basic sentence pattern: subject and
verb, or subject, verb, and complement. . . . Because English
has dropped the majority of case endings, the sense of the
sentence is peculiarly dependent upon logical order (dog bites
man; man bites dog). When the arrangement of the words or the
punctuation does not make this relationship clear,
communication is impeded. . . . The necessary basic kit is
comparatively simple. First, the writer needs to learn two
122
fundamental facts: (1) punctuation separates, and (2) internal
punctuation is used to show interruption or other irregularities
in the basic sentence pattern. . . . The next step would be to
learn about simple adjectives and adverbs and their normal
positions. These also would require no punctuation unless they
were multiplied.
(Yaggy, 1953:129.)
Yaggy’s references to separation, interruption, and irregularity really amount
to the same thing as preventing ungrammatical modification of the canonical
sentence because this is what these actions would accomplish. She also
seems to be hinting at recursiveness in mentioning something “multiplied.”
Unfortunately, Yaggy provided few examples to explicate her concepts, but
her approach makes it clear that she was suggesting the same basic idea: Use
the uncluttered canonical sentence as a starting point, then add the types of
details that one is likely to encounter in more complex sentences, such things
as multiple adjective modifiers, sentence adverbs like “however” and
“nevertheless,” along with “participial modifiers, appositives, and relative
clauses” and “the ubiquitous prepositional phrase” (Yaggy, 1953:130). But one
should prevent units that do not directly modify the canonical sentence
(relative clauses, appositions, and the like) from doing it by using the comma.
Clearly, Yaggy saw the comma as the only feature of punctuation
operating directly in relation to the canonical sentence. To complete the
picture, she offered some ideas for dealing with coordination, the semicolon,
123
and colon which are useful. Her approach could probably be described as
“structural” for she begins with the basic sentence as the primary structure,
looks at how various parts and pieces are attached to it, and shows how they
must be punctuated to preserve the schema of the sentence.
4.9 The New Paradigm
Walter J. Ong took a practical stand about writing. He announced it quite
clearly in the title of one of his writings: “Writing is a technology that
restructures thought” (Ong, 1992:293). Yes, he said, “writing (and especially
alphabetic writing) is a technology, calling for the use of tools and other
equipment” (Ong, 1992:299). When we use a tangible thing to accomplish an
action that we can’t do with our bodies alone, we are using tools, and tools
involve technology. It comes in many forms, affecting every one of our senses.
And the best technology of all, the most advanced, says Ong, is writing:
Writing, in the ordinary sense of a coded system of visible marks
enabling a writer to determine, in effect without limit, the exact
words and sequence of words that a reader will generate from a
given text, is the most momentous of all human technological
inventions. It is not a mere appendage or accessory to oral
speech. Because it moves speech drastically from the oral-aural
or voice-and-ear world to a new sensory world, that of vision,
124
writing transforms speech and thought as well.
(Ong, 1992:304.)
It should be hardly surprising to learn that one of Ong’s professors at St.
Louis University was Marshall McLuhan, whose famous slogan, “the medium
is the message,” dominated many discussions about the importance of form
for several decades but has recently been put aside. It may be time for its
revival. When it comes to writing, and including punctuation, form is at least
as important as content—and probably more important. The graphical form
of writing is the medium that delivers the message to us.
4.10 English Writing as a Graphic
Writing’s purpose is to use a language to communicate something, but it is a
machine that is essentially a graphic presentation. P. T. Smith et al. put it
this way:
The problem for the design of an efficient writing system is that
writing systems are essentially linear: a single sequence of
symbols must suffice to represent a complex hierarchy of
linguistic units. Phonemes, syllables, morphemes, words,
syntactic and semantic structures are all implicit in linguistic
output and may be inferred by a listener, but only a fraction of
this information can be conveniently transcribed by a simple
linear sequence of symbols.
(Smith et al., 1984:104.)
125
Like other machines, writing has an organized set of parts. The content of
language is, in abstract linguistic terms, phonology, morphology, syntax, and
semantics. But the form of writing is essentially a gestalt, and many of the
characteristics allied with the graphic aspects of gestalt psychology are
identifiable in it. The English written sentence is indeed the quintessential
graphic.
Stephen A. Bernhardt said it without equivocation: “Print is a graphic
medium; it displays its meanings in the spread of ink on page” (Bernhardt,
1993:168). Waller also anticipated this when he said that “’Text as diagram’
is a useful metaphor because it focuses us on the written-ness of text, mostly
ignored by linguists and those who study reading and learning” (Waller,
1982:137).
Written English obeys all the rules for graphics, especially those that
are related to gestalt psychology. To begin with, like other graphics, it has
the two basic parts: figure and ground. The ground for writing is the location
for it. Over time it has consisted of many different things and can still be any
one of them: a rock, a board, the wall of a cave or building, a sheet of glass,
the hide of an animal, a field of corn, a television screen, a computer monitor,
a sidewalk, droplets of water falling through space, a hologram, a sheet of
paper, and many more.
The figure, which is the basic content of the graphic, in English
consists of twenty-six minuscule graphemes:
126
abcdefghijklmnopqrstuvwxyz
The graphemes reside in the matrix, which has a vertical dimension and a
horizontal dimension and looks something like this:
vertical
|
|____________________
horizontal
This is a common L shaped graph form that is used frequently to show the
relationship of variables on two axes. There are technically no “spaces” in the
matrix, and there is no need to create any. Any gaps that we might perceive
are simply the ground showing through.
A very basic stretch of writing in English, then, consists of
minuscule graphemes arranged in various configurations in a
sequence on the ground in an orthogonal matrix. The sentence (or a
partial one) is a basic form found in writing. It organizes information
syntactically and has horizontal and vertical aspects. It has a definite
beginning but is potentially infinite because of recursion. It is displayed in an
orthogonal matrix. It might be easier right away to see the linear sentence as
a set of boxes into which we can add the vertical features.
□□□□□□□□□□□□□□□□
Since not all writing systems function in accord with horizontal and
vertical dimensions as English does, it would be advantageous for the
purpose of any future cross-linguistic study to describe what is occurring in
these locations in terms of the actions taking place rather than using the
names of the axes. Therefore, I will discuss English writing in accord with
127
that approach and not use the dimensional names. I will refer to the vertical
as the “static” axis and the horizontal as the “dynamic” axis. The terms
“static” and “dynamic” should also be more useful in describing what occurs
in writing systems which do not have clearly perceivable vertical and
horizontal axes. Giardino, coincidentally, in an article about trying to just
classify diagrams, uses the same terms, “static” and “dynamic,” to describe
their axes (Giardino, 2013: 243). Many of the orthogonal charts you see are
arranged that way, with the static (vertical) axis on the left and the dynamic
(horizontal) axis on the right, with data moving to the right on a line—much
the way a sentence does.
Seeing a sentence as a graphic is not a particularly new approach.
Scholars involved with artificial intelligence have been working on the
problem of mixing linguistic characteristics with graphic ideas for some time.
Shimojima provided a discussion of seven different ways in which linguistic
and graphic properties can be combined—without deciding which one was the
most viable for sentences, unfortunately (Shimojima, 2001). Steedman, in
trying to make syntax and phonology fit together, also talks about the rightangle shaped characteristics of the sentence. According to him, “Phrasal
intonation in English is frequently orthogonal to traditional notions of
surface syntactic structure” and he claims that Halliday (Halliday, 1967)
some time ago was already talking about this kind of dimensional view of the
sentence (Steedman, 2000:656). Steedman and Halliday, of course, were
128
probably talking about spoken rather than written English, but they were
both mentally conceiving the sentence in a more or less graphic way.
Roy Harris noticed the same thing in the sentence and saw that “two
variables are usually plotted orthogonally within the graphic space” (Harris,
1995: 123). But, unfortunately, he dismissed the possibility that a language
could reside in such a space:
Instead of a configuration of points related to the two axes, we
should find a series of sentences (i.e. arrays of alphabetical or
other scriptorial forms) displayed on the page in a pattern which
bears no relation at all to the co-variation of the values in
question. Nor would the value of any individual scriptorial form
depend on its occupying one particular position determined
by a calibration of horizontal and vertical axes on the page.
(Harris, 1995:123-124.)
Harris, apparently, was not able to see that the “figure” area consists of a line
of spaces, some of which are filled and some of which are empty.
Written English, then, is essentially an “orthogonal projection,” with
the horizontal axis representing the syntactic activities and the vertical axis
representing phonology, morphology, and some semantic features. Or, as the
dictionary says it, “a system of making engineering drawings showing two or
more views of an object at right angles to each other on a single drawing”
(Oxford American College Dictionary, 2002:964). Generally, the activities in
129
the static axis occur one space at a time as they affect individual linguistic
features. But the actions in the dynamic axis are spread over a number of
spaces in a line moving to the right in English.
Unfortunately, only a few people involved in fringe areas of linguistics
have discovered the orthogonal characteristics of written English. Some are
involved with things like speech recognition systems. Scheiber and Tau, who
were dealing with comma problems and surveying other studies, found that
“These studies, typically based on augmenting a Markovian language model
with duration or other prosodic cues as conditioning features, show that
prosody information is orthogonal to language model information . . . “
(Shieber and Tao, 2003:147). Daelemans et al., who were dealing with
natural language processing problems, also noted it:
On one strategy, information is partitioned between parental
nodes. You can, for example, inherit morphological properties
from node A and syntactic properties from node B, but no single
property can be inherited from more than one parent node. This
is known as orthogonal inheritance.
(Daelemans et al., 1992:209.)
Unfortunately, if you have a real interest in orthogonality, a random search
may lead you more to articles on Cartesian mathematics and discussions of X
and Y (two planes that intersect at a right angle) than to English writing
topics.
130
4.11 Gestalt and the Text
According to Lothar Spillman, in the Preface to Max Wertheimer’s On
Perceived Motion and Figural Organization, it was Wertheimer’s 1912 article
in the Zeitschrift für Psychologie on “Experimental Studies on Seeing Motion”
that began a flurried but relatively short study of what is now briefly known
as “gestalt,” which often translates as “form” (Wertheimer, 2012:ix).
Gaetano Kanizsa, however, believes that the term “gestalt” has
acquired too many ambiguities over time and ought to be translated as:
“organized structure,” as distinguished from “aggregate,” “heap,”
or simple “summation.” When it is appropriately translated, the
accent is on the concept of “organization” and of a “whole” that is
orderly, rule governed, nonrandom. This concept is opposed to
that of a merely, arbitrary, random and unstructured grouping.
But in addition to its being used to describe the product of a
process of organization, the term “gestalt” also indicates the
structural properties of the process itself. The term “organism”
could be used, but it too is inadequate because, in its most
frequent usage, it refers to a particular kind of gestalt (plant or
animal). Similarly, the use of “form” is inappropriate
because form is only one of the attributes of a gestalt.
(Kanizsa, 1979:56.)
131
To make matters even more complex, the term, which Wertheimer talked
about in terms of actual physical perception, became entangled with a variety
of psychological issues, muddying the waters even more. Curious slogans that
emerged, like “the whole is greater than the sum of the parts,” which Kanizsa
also takes issue with, did not help to clarify what gestalt is really about
(Kanizsa, 1979:61).
Several other German scholars joined the fray early, Kurt Koffka and
Wolfgang Köhler, but all of those involved early in getting the gestalt ideas
off the ground got sidetracked by the two world wars and the Jewish
diaspora. Wertheimer and Koffka died in the early 1940s and Köhler in 1967.
According to Roy Behrens, the gestalt “influence in the field of psychology is
unobtrusive in the sense that its findings have all been absorbed by more
recent viewpoints,” and no major names have been associated with it
recently, with the possible exception of Rudolf Arnheim, Gyögy Kepes, and
Donis A. Dondis (Behrens, 1998:302). Otherwise, the result has been
something like a long hiatus in discussions of the gestalt theories in how
visual perception works. Only here and there throughout the twentieth
century can one find an article devoted to what were the clear and fascinating
facets of perception that the gestaltists were beginning to reveal as so
important in how we perceive the world around us. Also, there is often some
disagreement among the writers on the number of basic gestalt features, so
132
what amounts to less than a half dozen features in one source comes close to
a full dozen in another.
Perhaps it is Chang, Wilson, and Dooley who have collected the
greatest number of gestalt “laws” from a variety of other sources and
compiled them in an article. They found eleven of them, as follows:
Law of Balance/Symmetry: suggests that design should aim
for symmetry, balance, and proportion to promote the sense of
equilibrium
Law of Continuation: human eyes are inclined to follow an
object’s direction; continuation is the eye’s instinctive action
Law of Closure: Our brains instinctively enclose a space by
completing an outline and ignoring gaps in the visual concept,
thus completing unfinished forms
Law of Figure-Ground: it is natural for humans to distinguish
between a foreground and background
Law of Focal Point: the purpose of the focal point is to attract
the viewer’s attention and get them to look further. It is the
most important area compared to other parts of the visual area
Law of Isomorphic Correspondence: the kind of audience
involved must be taken into consideration so as not to mislead or
confuse them with the designs used
Law of Prägnanz (Good Form): good form is a simple design,
133
such as a simple circle, a symmetrical layout, or something that
tends to impress the observer
Law of Proximity: elements close to each other are seen as
related, and those far apart are less related or unrelated, so
items near each other are seen as part of a group
Law of Similarity: similar objects will be viewed as part of the
same group, which can be used to capture attention and focus
on key points
Law of Simplicity: elements and objects should be arranged in
a simple manner; complex design may cause misunderstanding
Law of Unity/Harmony: related objects should appear in the
same form so they seem to belong together .
(Chang et al., 2003-2004:4-12.)
[I took the liberty of shortening the descriptions of the “laws” but tried to
preserve the key factors in the explanations.] Their article was actually
aimed at someone intending to prepare a multimedia presentation, but it
should be evident that virtually all of the laws mentioned could apply to
every written text or any kind of visual medium.
The degree to which writing obeys these “laws” should also be obvious.
Writing is understood to contain meanings, so even the smallest thing that
you put on a page of print will be scrutinized to discover its meaning. Writing
embodies the laws in a number of ways. Figure and ground we have noted
already. Capital letters to start and terminal punctuation to end a sentence
134
operate with continuation and closure. The line on which the text exists is
part of the symmetry and linearity, while words that are joined by a
hyphen work with the idea of proximity. Some of the laws, of course, may
relate more to how style is used and overlaid on a text, things such as the
font and other typographic features the writer may have chosen to use. But a
brief study of the “laws” and a look at a line of text will quickly confirm that
all of them are present in some way.
Michael Twyman, in “A Schema for the study of Graphic Language,”
notes the inevitable linearity of writing, even if it occurred in a circle in some
texts in Minoan from about 1700 BCE (Twyman, 1979:123). Turnbull and
Baird note that the eye tends to move across an area from left to right and
prefers horizontal movement (Turnbull and Baird, 1975:167-168). Donis A.
Dondis, in A Primer of Visual Literacy, in what could be a comment about a
sentence, says that:
line, because of its nature, has enormous energy. It is never
static; it is the restless, probing, visual element of the sketch.
Line, wherever it is employed, is the essential tool of
previsualization, the means for presenting in palpable form that
which does not exist yet, except in the imagination. . . . Yet for
all its looseness and freedom, line is not vague; it is decisive; it
has direction and purpose, it is going somewhere, it is doing
something definitive. . . . Line is also a tool for notation
135
systems, writing . . . [and] symbol systems in which line is the
most important element.
(Dondis, 1973:43.)
Even the least noticed of the static (vertical) punctuation features gets a
boost from a gestaltist. Rudolf Arnheim extols “the attractive simplicity of
static concepts, which pick out some one characteristic state of an object or
movement and let it stand for the whole” (Arnheim, 1969:178-179). Arnheim
could easily be talking about a capital letter—or an apostrophe. Lest we
overlook the minutiae, Martin Solomon, in “The Power of Punctuation,” says
that “The sensitive application of punctuation even in the most commonplace
unit changes the entire feeling of a design” ( Solomon, 2004:287).
George Gerbner, in “The Interaction Model: Perception and
Communication,” perhaps unknowingly echoed the gestaltist mantra when he
said that:
Communication is communicating. It is a pattern of doings, a
process. This means that it is a flow of events so interrelated
that one act in the series derives part of its significance from all
the other acts, and can be fully grasped only in the light of the
total pattern.
(Gerbner, 1960:6.)
For some reason, it almost sounds like Gerbner is saying that the whole is
greater than the sum of the parts.
Visual language, then, operates in a significantly different context
than spoken language, which is arbitrary, random, often unstructured, and
136
has its own set of rules. It may be possible to do some kind of comparisons of
the content of the two, but the forms are different enough so that casual
comparisons are liable to misrepresent how something is intended as well as
what is intended. If you hear a voice while you are reading (it is called subvocalizing), for example, the voice you are hearing is your own, not that of the
writer, who had to obey only the visual rules while creating the text. Writing,
of course, is based on a more neutral standard with conventional rules, not
necessarily the writer’s dialect.
4.12 Locating Punctuation
Most people have the idea that the role of punctuation in a written text is
rather minor and consists of a few leftover things that have to be added to an
otherwise clear and well-organized piece of writing. Here is a working
definition for punctuation:
Punctuation is a system of symbols and graphic features used to
protect the integrity of the sentence in the English writing
system and make it visible.
A symbol is usually defined as an object that has no intrinsic meaning but
acquires it by use over time by human beings. The definition will be used for
both the static and the dynamic axes, though it will be applied differently.
But there are a few things we need to clarify about symbols. Our
interest is only in symbols that have a direct influence on one or more of the
137
linguistic units, the semantics, syntax, morphology, or phonology of the
English writing system. There are many other symbols that might be found
in a written text. Unicode, a character encoding standard used in
programing, also provides lists of symbols that are labeled as punctuation but
really are not according to our definition (for Unicode symbols, see
http://www.fileformat.info/info/unicode/category/Po/list.htm). Most of them
do not modify any linguistic units. Also, symbols such as %, @, and # are
included, but these and many other symbols in Unicode are just synonyms for
words like “percent,” “at,” and “number.”
On the other hand, some symbols used to indicate abbreviations might
fit into our definition because they usually show that a longer word has been
shortened in some conventional way, thus indicating that a modification has
taken place, one of the two uses of the period. “Mister” becomes “Mr.” and
Miss” becomes “Ms.” (However, to confuse the issue, some publications have
even eliminated the punctuation with abbreviations in some contexts).
Acronyms are also a questionable category.
When the punctuation is omitted and they are pronounceable, they
move to the category of words in English. Thus, NATO is pronounceable as a
word. If it were up to Robert Bringhurst, author of The Elements of
Typographic Style, some of these acronyms that are pronounced should even
shed their capitals:
138
When a writer accepts them fully into her speech and urges
readers to do likewise, it is time for the typographer to accept
them into the common speech of typography by setting them in
lower case: Unesco, Ascii (or ascii) and Fortran. Other acronym
words, such as laser and radar, have long since traveled the
same road.
(Bringhurst, 2005:49.)
Language is changing all the time. Just about the time we start to believe the
lexicon is rather stable, some kind of alteration or borrowing is liable to
occur.
When you look at a typical page of written English, you are likely to
see more than the twenty-six minuscule graphemes. There are many
arguments about which of the items one sees on a page of writing should
really be classified as punctuation. But using our definition and the list of
items we classified as punctuation, everything else you might see that is
added to or included with the twenty-six minuscule graphemes is actually
punctuation. This can be demonstrated. Below is a sample paragraph which
includes all the items of punctuation from the reference works mentioned
earlier. They have not been preselected or limited to just certain ones as was
done with virtually all of the studies discussed in Chapter 2.
139
4.13 The Sample Paragraph with Punctuation
This sentence is simple. But, when you start getting into some depth and
want to add a variety of things to your sentence, it gets complicated. An
appositional structure, like this one, may require several commas around it to
make sure you don’t end up with a noun phrase with a different meaning.
That goes for a relative clause as well, which may or may not modify the
noun it’s near. Word modifications, such as a nice, new adjective added to a
subject noun, are not usually a problem; but transformations—that is, really
sentences that have been truncated in some way, (e.g., even a relative clause
without a relative pronoun, for example)—can cause considerable confusion
when the punctuation is missing. As someone once said: “. . . a sentence is a
complete idea!” [Did I get that right?]
4.14 The Sample Paragraph without Punctuation
thissentenceissimplebutwhenyoustartgettingintosomedepthandwantoaddava
rietyofthingstoyoursentenceitgetscomplicatedanappositionalstructurelikethis
onemayrequireseveralcommasaroundittomakesureyoudontendupwithanounp
hrasewithadifferentmeaningthatgoesforarelativeclauseaswellwhichmayorma
ynotmodifythethissentenceissimplebutwhenyoustartgettingintosomedepthan
dwanttoaddanounitsnearwordmodificationssuchasanicenewadjectiveaddedtoa
subjectnounarenotusuallyaproblembuttransformationsthatisreallysentencest
hathavebeentruncatedinsomewayegevenarelativeclausewithoutarelativepron
140
ounforexamplecancauseconsiderableconfusionwhenthepunctuationismissinga
ssomeoneoncesaidasentenceisacompleteideadidigetthatright
4.15 The Importance of Space
The lack of spaces in this last paragraph sends us back at least several
thousand years in the history of writing to scriptura continua and makes it
difficult for a modern reader to locate the words. The argument for including
spaces as an “official” part of the punctuation of a written graphic is rather
strong and has been mentioned by a number of scholars over the years.
Mountford, in discussing ways to sort out writing systems, considers
including “punctive” items such as “white space” and “word space” among
other criteria (Mountford, 1980:229). Waller claims that:
Punctuation is the single aspect of written language, for which
grammatical rules exist, that does not represent words
themselves but the spaces between them. It is, then, an
organizational system at the micro-text level functioning in
much the same way as typographical signals and the use of
space at the macro-text level.
(Waller, 1980:245.)
Southhall says it briefly: “Written language contains elements which are not
alphabetic or numeric characters: punctuation signs and space” (Southhall,
1984:83). Nunberg says that punctuation “must be considered together with a
variety of other graphical features in the text, including font- and face-
141
alterations, capitalization, indentation and spacing, all of which can be used
to the same sorts of purposes” (Nunberg, 1990:17).
Richard L. Venezky (1970:47) showed his support for the idea by
repeating some lines directly from W. Nelson Francis:
Taking inventory of the segmental graphemes of standard
English writing, we find that there are thirty-seven of them,
which can be classified in two groups:
(a) Twenty-six letters of the alphabet <abc . . . z>
(b) Eleven marks of punctuation <, ; : . ? ! ‘ – — “ (>
In addition, we must include space, a sort of zero grapheme.
(Francis, 1958:436.)
This judgment by Francis seems prescient. (I can accept the twenty-six
minuscule letters that Francis mentions but may have more to say about a
classification involving graphemes later.) In talking about space as a
grapheme, however, Venezky and Francis are in fact giving it an equal place
in the graphic scheme. But a space is not a grapheme. Graphemes are the
representatives of phonemes. Spaces are the ground showing through.
Just how important are spaces? Robert D. Stevick rued the fact that
some additional and meaningful spaces were eliminated from some Old
English manuscripts when new print editions were produced:
That is, the [new] editions disregard the spacings in the
manuscript, assuming them to be irrelevant, if not arbitrary or
142
capricious. Yet these were produced by Anglo-Saxons who, to
their credit and to our good fortune, had not embraced
“canonical word separation” and instead used spacing to record
something more than lexical demarcations.”
(Stevick, 2004:3.)
Stevick believes the purposeful spacing in the Anglo-Saxon manuscripts was
related to prosodic features that the writers could not show in any other way
(Stevick, 2004:11). The Anglo-Saxon scribes were, in effect, using space as
meaningful punctuation. Perhaps inspired by those Anglo-Saxon scribes, a
few modern linguists have made an effort to improve readability by altering
the spaces in texts using some computer programs to locate phrasal
boundaries. They were able to prove that “isolating major phrases within
extra spaces facilitates reading, especially among poor readers” (Bever et al.,
1990:83).
Readers of some non-alphabetic languages may actually be more likely
to believe that space is important in their texts than English readers and
writers. According to Roy Harris, space “is in a sense ‘built into’ Chinese
writing, since in most cases the individual character identifies a word”
(Harris, 1995:171). An extreme case, he says, is Classical Tibetan, “where
syllable division, indicated by punctuation marks, becomes the organizing
principle of the whole arrangement of character” (Harris, 1995:171). Most
people looking at English writing, however, probably don’t have a clear idea
143
how space fits into the scheme of things, and it remains more or less invisible
to them.
John Lennard, in his article titled “In/visible Punctuation,” suggests
that the failure to see space as related to punctuation is an act of denial:
The most obvious and important example of invisibility arising
from such denial is spaces of punctuation, and the problem
begins etymologically. ‘Punctuation’ derives from Latin punctus,
a participle of pungō, ‘to puncture, prick (a hole),’ once a literal
piercing of parchment with a sharpened point, most probably in
tallies, but transferring to use of a stylus on wax. Most modern
definitions of the English word . . . consequently insist that
‘punctuation’ is synonymous with ‘punctuation marks,’ i.e., that
it comprises only points and other non-alphabetic marks
interspersed among words. This ignores the Latin extension of
the term from a point in space to a point in time, usage reflected
in modern English ‘to punctuate’ (inter alia, to “interrupt at
intervals: intersperse with” . . .), in ‘punctual’ and its cognates,
and in the common compound noun ‘punctuation marks’ . . .
which would be needless if there were no other kinds of
punctuation to distinguish.
(Lennard, 2011:123.)
What Lennard is suggesting is that, if there were a class for features of
punctuation, some of them would be “marks” and some would not because the
144
word “marks” seems to modify the word “punctuation.” (Thus, there should be
some that are not marks. So we need a better word to describe the features.
We will talk about that in Chapter 5.) Lennard continues his argument by
noting that anyone who has experienced the unspaced early writing called
scriptura continua (like our unpunctuated sample paragraph above) should
be able to tell that modern writing is clearly punctuated using space in many
places, “between words, in conjunction with full-stop and capital letter
between sentences,” and just about everywhere else (Lennard, 2011:124).
Roy Harris, in an appendix to Signs of Writing, refers to “word
division” when he means “spaces,” saying that “Word division itself, although
not always recognized as a form of punctuation in modern manuals
(presumably, again, because a blank space is not counted as a ‘mark’), is one
of the early aids to text processing supplied by the writer” (Harris, 1995:171).
So he confirms Lennard’s view of the confusion punctuation faces because of
the terminology.
Martin Solomon, in an otherwise excellent article on the typographical
uses of punctuation, seems to have gotten sidetracked by the terminology in a
similar way. Because of his use of “punctuation mark” so often in his article,
he apparently began to believe that “underscores are not punctuation” simply
because they are not “marks” (Solomon, 2004:286). Solomon apparently was
involved in what the gestaltists called ‘fixation”:
145
Gestaltist psychologists were especially concerned with
situations where an individual misinterprets the situation or
fails to see the true structure. The person is described as
“fixated” on an inappropriate interpretation of the problem.
(Dominowski and Dallob, 1995: 45.)
Lennard already noted a similar problem. A person who only thinks of
punctuation as “marks” is unable to see all the other features that are doing
the same work.
The prize for talking about the invisible characteristics of writing goes
to Paul Saenger, the author of Space Between Words: The Origins of Silent
Reading. It seems hard to believe, but Saenger reveals that spaces between
words had to wait until the twelfth century to become a regular thing
(Saenger, 1997:44). Apparently, because they are always out of sight, spaces
to many modern writers on the subject of punctuation are also out of mind.
4.16 Putting the Spaces back into the Sample Paragraph
this sentence is simple but when you start getting into some depth and want
to add a variety of things to your sentence it gets complicated an appositional
structure like this one may require several commas around it to make sure
you dont end up with a noun phrase with a different meaning that goes for a
relative clause as well which may or may not modify the noun its near word
modifications such as a nice new adjective added to a subject noun are not
146
usually a problem but transformations that is really sentences that have
been truncated in some way e g even a relative clause without a relative
pronoun for example can cause considerable confusion when the punctuation
is missing as someone once said a sentence is a complete idea did i get that
right
The spaces obviously make the sample paragraph more readable and
probably confirm their importance as a feature involved with punctuation.
They are the essential static (vertical) feature which modifies all of the
linguistic features in the matrix by preventing their collision. The writer
using spaces has made decisions about how the features will be seen and
understood, mostly words, of course. Even with the spaces, though, we still do
not always have a clear understanding of how the pronunciation of individual
graphemes might be altered, which words might be compounds or
abbreviations, and which words constitute sentences or sentence partials.
More punctuation might be needed in the future to accomplish those things.
Another possible problem with the word “space” in talking about a
written text is that it might suggest that a writer has made a conscious
decision to create one, but the spaces will really be there whether or not the
writer uses them. We are dealing with a graphic which has its own
characteristics, a matrix that is predetermined. Space is primitive
punctuation. It was there when there was no text. A better mind set would be
that the writer may only decide not to fill a space. And perhaps an even
147
better way to deal with this might be to use another term that is sometimes
part of the vocabulary of graphics. The line of writing, in effect, is made of
“cells,” and a writer can decide one of three things: to fill a cell with a
grapheme, some punctuation, or leave it empty.
4.17 Phonological Punctuation
There should be no question that punctuation relates to linguistic features. It
is used to modify every kind of unit in writing, one cell at a time. The
phonological punctuation includes capitals, diacriticals, and graphic devices
for calling attention to one or more cells. We might also think of them as a
class of “single cell” punctuation because they only modify one cell at a time.
A diacritic might be the only one that seldom modifies more than one cell.
Each type of diacritic is usually involved with only one particular grapheme.
Capitals can be used on one or more cells. The use of a capital to start a
sentence may seem like a good idea, but it could also be a source of
ambiguity.
4.18 Morphological Punctuation
Morphological punctuation is rarer than the phonological. The hyphen, the
apostrophe, and the removal of an empty cell are the most common. The
hyphen is occasionally used to remove ambiguity; for example, it can be used
to prevent words like “co-op” from accidentally being read as a place where
148
chickens are kept. Individual words are sometimes linked to create a new
compound like a “once-in-a-lifetime” offer. The hyphen is also used to
separate syllables when a word needs to be broken at the end of a line of
writing. This use of the ordinary hyphen for doing this, however, is also
problematic, and it can cause trouble in the same way as using a capital to
start a sentence.
The apostrophe has several uses. One is to indicate a missing letter in
a contraction. You eliminate one of the usual graphemes in a word and put an
apostrophe in its cell instead. Another use is to show possession. The singular
possessive is not usually a problem, but plural possessives or words ending in
“s” create some arguments among editors, mostly about the appearance, and
there is a tendency to omit a dangling letter. Most of such arguments about
punctuation have to do with typographical style rather than linguistics.
A new word is sometimes created by skipping the space between two
words that are so often sequenced that they come to be thought of as a single
conception. Something that happens around the world a lot can be considered
a “worldwide” event, thus creating a new adjective and saving some space.
The period used to indicate an abbreviated word might be added here. It
actually does modify a morphological unit by shortening it and is usually not
likely to be confused with a sentence period because of its location (see Mr.
and Ms. above).
149
4.19 The Sample Paragraph with only Phonological and
Morphological Punctuation now Added
This sentence is simple But when you start getting into some depth and want
to add a variety of things to your sentence it gets complicated An appositional
structure like this one may require several commas around it to make sure
you don’t end up with a noun phrase with a different meaning That goes for a
relative clause as well which may or may not modify the noun it’s near Word
modifications such as a nice new adjective added to a subject noun are not
usually a problem but transformations that is really sentences that have
been truncated in some way e. g. even a relative clause without a relative
pronoun for example can cause considerable confusion when the punctuation
is missing As someone once said a sentence is a complete idea Did I get that
right
The paragraph is beginning to look much more normal, but all the
punctuation for the dynamic axis is still missing. Some of the sentences can
be discerned, but some are confused or remain mysterious. We need still some
dynamic and sentence level punctuation and interpolations. [You will find
them in the original version of this paragraph, section 4.13 above .]
4.20 Static Attention Devices
Punctuation such as underlining, italics, bold, or a change in a font or its size
usually, but not always, involves a row of cells. Typographic practices often
150
come into play with these features. For example, should someone underline
an empty cell? No matter how hard you try, however, you can’t italicize an
empty cell or make it bold. So each phonological feature has its own
characteristics, and only individual cells are affected. Each one creates a new
meaning when static punctuation is applied and modifies an existing
grapheme.
4.21 Punctuation and Markedness
From studies of markedness, we know that a change in form amounts to a
change in meaning. All of the symbols seen as punctuation on a page of
written English involve markedness. They embody the characteristics that
are talked about in discussions of markedness. One reason they may not be
understood as such is that they are more visible. Thus, they may be perceived
in a way different from other items of markedness, which are usually
discussed in the context of speech and often not noticeable unless someone
calls attention to them. Punctuation is instead silently immersed in a
graphical context—barely seen and certainly not heard.
Markedness, after all, is about difference in just a small way, not a
large way. Because it is always present, it is easy to overlook what is
happening with some punctuation. It becomes part of the general appearance
of the page and is not consciously noticed for its influence on the language. It
seems to be creating some kind of new “order” but not exhibiting an overt
151
linguistic difference on a text. (This may be why many younger people do not
pay much attention to them.) Indeed, many of the studies of punctuation
dwell on the idea of order rather than linguistic modification and miss the
point.
The main idea behind markedness, according to Moravcsik and Wirth,
is one of correlation:
The domain of the theory is in all cases pairs of opposing
language-structural entities that exhibit an asymmetrical
relationship in more than one respect. The central claim is that
the various tests that demonstrate the symmetry between the
two members of the opposition will have converging results:
Once one of the two members has been shown to be marked by
one criterion—let us say, it has been shown to be structurally
more complex than the other, or paradigmatically poorer, or
more restricted in its distribution—all other relevant tests
will also converge to select that entity as the marked member of
the opposition.
(Moravcsik and Wirth, 1986:3.)
One set of those perhaps “relevant tests” is noted by Edna Andrews
and comes from a domain other than linguistics. Andrews makes the point
that there is a significant connection between form and meaning, that
opposition is a key factor in markedness, and that the only way to ultimately
152
discover it is by way of “form.” The first difference noted in markedness then
is “formal,” and the second is “meaningful” (Andrews, 1990:95).
Andrews introduces nine axioms from set theory to bear on any
situation involving markedness. The first is perhaps the most obvious for
punctuation:
Axiom 1: A Difference in Form Signifies a Difference in Meaning
The very embodiment of punctuation is form. Limited visibility is its forte.
Each modification alters a linguistic form in a simple but meaningful way.
Axiom 2: Every Set Can Be Well-Ordered
If we consider all of them as a group, punctuation items appear quite distinct
in how they alter the appearance from one linguistic group to another. They
are well-ordered in that sense. And those that actively modify phonologically
or semantically are quite distinct from those that function morphologically or
syntactically.
It is likely that we could find comparative values in the remaining
axioms Andrews discusses, but some seem to require greater numbers for
comparison than punctuation has to offer for a test, number 4, for example:
Axiom 4: For Any Two Sets, There Exists a Set
That They Both Belong To
There just aren’t enough items in the pool of punctuation to make a good test
here, it seems. An exception might be the punctuation involved with syntax,
with the comma representing one set and interpolations items representing
153
another, although the interpolations don’t participate in the syntax of the
basic sentence (excepting quotations, perhaps). The final axiom Andrews
offers, however, rings true:
Axiom 9: There Exists a Set Which Has No Elements.
(Andrews, 1990:97-109.)
Punctuation clearly has that one: space.
With just a few items, T. Givon, in Syntax: A Functional-Typological
Introduction, Volume II, also offers a simple test for markedness, which
together will confirm our belief in punctuation as markedness:
(a) Structural complexity: The marked structure tends to
be more complex—or larger—than the corresponding
unmarked ones.
(b) Frequency distribution: The marked category (figure)
tends to be less frequent, thus cognitively more salient, than
the corresponding unmarked one (ground).
(c) Cognitive complexity: The marked category tends to be
cognitively more complex—in terms of attention, mental effort
or processing time—than the unmarked one.
Perhaps in anticipation of this study of punctuation, Givon has brought up
the importance of figure and ground several times here. He states that (b)
above “is intimately associated with the cognitive phenomenon of figureground relations. Most perceptual and cognitive distinctions tend to pair up
154
so that the salient, important, ‘figure’ is less frequent” (Givon, 1990:947-948).
It is likely that Givon was thinking of the figure-ground relationship in a
metaphoric way in relation to speech rather in an actual graphic, but we will
accept his words as meaningful in the application to a graphic context as
well.
Sometimes less is actually more, especially in graphic terms. According
to Edward R. Tufte, the best design strategy involves “the smallest effective
difference”:
Make all visual distinctions as subtle as possible,
but still clear and effective.
Tufte invokes Occam’s razor (“what can be done with fewer is done in vain
with more”), which is relevant to “nearly every display of data” (Tufte, 1997:
73). We know that the ground even has a role in the more effective use of
features. Some punctuation items take less than a whole space and allow the
ground to show through. Furthermore, says Tufte, “Minimal contrasts of the
secondary elements (figure) relative to the negative space (ground) will tend
to produce a visual hierarchy, with layers of inactive background, calm
secondary structure, and notable content” (Tufte, 1997:74). Perhaps the only
items that fail to comply with this standard are some of the semantic ones,
with their purposeful and sometimes exaggerated use of different fonts, sizes,
bold, underlining, and italics, which, Tufte suggests, tend to activate the
ground, something that may be a distraction on a page of writing. But,
155
practically, their aim is to get our attention, and they certainly do, sometimes
annoyingly.
“Confusion and clutter,” according to Tufte, “are failures of design, not
attributes of information” (Tufte, 1990:53). The goal is to find a strategy that
adds information without adding complication. Again, punctuation complies.
Two features suggested by Tufte occur in punctuation: layering and
separation. The semantic and phonological punctuation simply add a feature
to the existing grapheme when they add another meaning. This amounts to
layering. Syntactic and morphemic punctuation use space as a feature,
creating a visual distance that in a graphic way also represents the
grammatical distance they are creating. This amounts to separation. (Indeed,
“separation” is a word often mentioned in discussions of punctuation, but it is
usually meant in a different way.) At about this time, we begin to wonder
how, without a trained graphic artist on the payroll, the many writers in
English over the past thousand years managed to devise such an efficient
visual system for displaying the symbols needed to make our written
communication as clear and as well organized as it is.
But perhaps the “smallest difference” is not always effective for some
people and may be the reason that punctuation is nearly invisible to some
younger readers, who may because of their inexperience be having trouble
more with the words and syntax than the small symbols for punctuation. But
Mina Shaughnessy found that adults classified as “basic writers” largely used
156
only commas and periods and that “they do not perceive the written sentence
as something that can be broken into or added onto for purposes of
elaboration, modification, or side comment” (Shaughnessy, 1977:28). For all
practical purposes, nothing other than terminal punctuation items seemed to
exist for these students. And why might they not see them? “The small marks
of punctuation, after all, don’t look very important. They don’t seem to say
much either” (Shaughnessy, 1977:27). The problem, however, seems to be less
about the lack of knowledge about punctuation than a dearth of knowledge
about the sentence and its capabilities, but they tend to go together.
Edwin Battistella, in an article titled “Marked and Unmarked
Punctuation Signs in English,” tried to find a way to compare items of
punctuation with each other to determine how they related in terms of
markedness (Battistella,1996). In doing so, he also avoided the term
“punctuation mark” because he wanted to use the concept of “zero” in his
comparison and knew it was in no way a “mark.”
His analysis is puzzling, however, because he was comparing items of
punctuation to each other and not to the linguistic features they represent (in
the case of the phonological, morphological, and semantic features) or the
ones involved with syntax in some way. Clearly, items of punctuation are not
in any way hierarchic with one another in the graphic matrix. Each has a
different purpose, a purpose which is related to one of the linguistic
categories. They are a system, a set.
157
Battistella’s final lines in the article revealed his uncertainty about the
endeavor: “It might well be that the features posited on the first pass through
the punctuation system are not correct—in fact, it would be quite miraculous
if they were” (Battistella,1996:246). I am likely to agree with him. (In an
earlier work, titled “Notes on the Sign Structure of English Punctuation”
(1993), Battistella looked for a way to connect punctuation with Peirce’s
system of signs, but he was only working with a few dynamic items and
posited some ideas without arriving at a conclusion.)
4.22 Phonemes and Graphemes
David Crystal describes the “-eme” suffix used in linguistics this way:
An ‘emic’ approach . . . takes full account of FUNCTIONAL
relationships, setting up a CLOSED system of CONTRASTIVE
UNITS as the basis of a DESCRIPTION. Emic is in fact derived
from such terms as PHONEME and MORPHEME, where –eme
refers to the minimal DISTINCTIVE units involved.
(Crystal, 1997:134-135.)
There are phonemes and morphemes, then. Both have been around for some
time. But the term grapheme is somewhat newer. It is the term that is used
to represent the phoneme in written form. But not everyone is happy with the
term because it has also sometimes been used to describe features other than
phonemes that are represented as characters in a written text.
158
Peter Daniels posed the question in his article: “Is a structural
graphemics possible?” Daniels aired all the various ways in which the term
“grapheme” has been applied, and the result is clearly confusing. Too many
people have tried to include too many items in the category and sometimes
debased it. He is correct in bringing this out and criticizing those who
misapply the term. In a later work, The Handbook of Linguistics, he makes
his displeasure clearer:
writing systems do not work like linguistic systems; there is no
“emic” level, and the popular term grapheme is misleading. For
instance, many alphabets use a pairing of symbols—capitals and
lowercase, majuscule and minuscule—that has no equivalent in
sound systems.
(Daniels, 2001:66.)
In the earlier work, Daniels does mention the work of Ernst Pulgram, who
went to great trouble to describe the family resemblance between phonemes
and graphemes in his “Phoneme and Grapheme: A Parallel” (1951). Pulgram,
according to Daniels, discusses “the parallels between phonemes and letters of
the alphabet; but even if we restrict the discussion to English, whatever the
concept of grapheme may cover, it must include more than letters of the
alphabet. Wouldn’t dollar signs be graphemes? Numbers? Punctuation
marks?” (Daniels, 1992:531)
At this point, it is clear that Daniels has himself crossed a red line.
When he tries to include majuscules, dollar signs, and punctuation marks in
159
the category of graphemes, he is doing the same thing as those he earlier
criticized: adding irrelevant features to the category of graphemes. My guess
is that Pulgram did an adequate job in classifying and comparing graphemes
and phonemes in a parallel way. Here is what he offered for graphemes:
1 The smallest distinctive visual units of an alphabet are its
graphemes.
2 A grapheme is a class of written characters pertaining to one
alphabet.
3 The hic et nunc [here and now] written realization of a
grapheme is a written alphabetic character or graph.
4 The number of graphemes in each alphabet must be limited,
the number of graphs cannot be.
5 By definition, all graphs identifiable as members of one
grapheme are its allographs.
6 The graphic shape of an allograph is dependent on its producer
and on its graphic surroundings.
7 The graphs which are not immediately and correctly
identifiable as belonging to a certain grapheme when occurring
in isolation, may be identified through their meaningful position
in a context.
8 Alphabets are subject to graphemic change and substitution.
160
9 The number, kind, and distribution of graphemes varies from
alphabet to alphabet. (Pulgram, 1951:15-16.)
Thus, only those items that were phonemes originally can also be considered
when looking for graphemes. Certainly, majuscules, dollar signs, and
punctuation marks were never included in Pulgram’s classifications—but
neither were spaces. Furthermore, some of Pulgram’s nine axioms can also be
used for defining a new term for punctuation as well, simply by changing a
few letters. (See Chapter 5.)
Leslie Henderson also examined the confusion about graphemes and
found at least three distinct uses. He ultimately seems to have agreed with
Pulgram’s approach, calling the grapheme “the minimal distinctive unit of a
writing system” (Henderson, 1985:146). This certainly describes the twentysix minuscule letters.
4.23 Punctuating the Dynamic Axis: the Comma
Baldwin and Coady, in their article titled “Psycholinguistic approaches to a
theory of punctuation,” found “that the rules of English punctuation
established with traditional grammar are empty conventions, which neither
predict nor explain reading behaviors involving punctuation” (Baldwin and
Coady, 1978:375). They determined instead that English punctuation was
actually a “cue system” . . . “an orthographic device which signals syntactic
patterns to the reader” (Baldwin and Coady, 1978:363-364). They performed
some experiments with twenty fifth grade students and twenty linguistics
161
graduate students, the basis for which was to see if the canonical sentence
and punctuation had any influence on their reading efficiency. They did make
a difference:
The results of the present research support the notion that
individual marks of punctuation exert a variable influence upon
sentence comprehension. Moreover, the canonical-noncanonical
distinction appears to broadly define the grammatical conditions
under which punctuation cues are redundant or critical aspects
of the visual display. When sentences are noncanonical,
punctuation seems essential in arriving at appropriate syntactic
analyses. When sentences are canonical, punctuation appears
merely to reiterate grammatical information already provided
by word order.
(Baldwin and Coady, 1978:370-371.)
The graduate students did adjust to punctuation peculiarities better than
the fifth graders, but the results showed that the canonical sentence had a
serious influence on comprehension and that punctuation played a part.
A few years later, Mohan R. Limaye used the idea of the canonical
sentence as part of a more systematic approach to punctuation by pointing
out that marked work order often requires punctuation (Limaye, 1983: 29).
By “marked” he meant noncanonical word order, and he showed that a
sentence that started with the SVO pattern needed no punctuation even after
other structures were added to it at the end. But when the SVO sentence was
162
delayed and the alternate structures put first, punctuation was necessary.
Essentially, his approach was similar to that of Baldwin and Coady.
Unfortunately, most of the examples used in the studies were rather
short sentences, and none of these scholars pursued the idea of the influence
of the canonical sentence on punctuation any more, but they made the point
that the canonical sentence, rather than the textbook rules, clearly had
something to do with the reason for punctuation.
According to Nathan Knobler, in The Visual Dialogue,
Communication is a transfer of information or ideas from a
source to a receiver. Some vehicle or medium is required for this
exchange. We usually refer to this vehicle as a “language.” As
communication becomes more complex, as the descriptions
become more precise, the language must be developed to
represent the specific information and express it. This process
requires a language of many individual symbols and a
systematic means for combining them into significant and
understandable relationships
(Knobler, 1966:33-34.)
In the next step, in English writing, we use graphemes, which are really
“signs” (rather than symbols) that represent phonemes and thus possess
some kind of meaning already. When language gets turned into writing, says
Knobler, “a combination of lines or marks” is used to form words like “CAT,”
which a seasoned reader will ultimately perceive as a whole word rather than
163
individual graphemes (Knobler, 1966:34). In the next step, moving into
syntax, according to Knobler,:
the sentence, “THE CAT IS ON THE FENCE,” is a grouping of
separate elements which the reader has learned to combine in a
particular way. If the accustomed pattern is broken, even
though the individual elements are unchanged, confusion can
result:
THEC ATIS ONTH EF ENCE
CAT FENCE ON IS THE.
Both of the above letter groupings are confusing, for they do not
fall into a pattern which has a familiar appearance.
(Knobler, 1966:35.)
In this digression into talking about language as an introduction to talking
about art, Knobler makes a significant point about written communication. It
behaves in a way similar to any visual communication where perception is
concerned and this provides a clue to why punctuation is important. It is
what we see and how we understand the relationship of the visual items that
is in control.
If all sentences were short and simple, there would probably be no
need for much dynamic punctuation. Static punctuation would probably take
care of most of the phonological, morphological, and other anomalies a
language is privy to. It is in syntax, when a sentence gets complex, that
164
dynamic punctuation is required. Just as the displacement of cells in the
ground in the short sentence
THEC ATIS ONTH EF ENCE
makes it hard to read, the addition of words and phrases to a simple sentence
can potentially create confusion because the relationships will not be
apparent without some system for showing them visually.
Beginning in the 1960s, a number of scholars became interested in how
sentence parsing worked. They devised a number of strategies on the way to
finding out how sentences managed to arrive at some kind of syntactic
completion. One persistent question was the relationship of semantics and
syntax. Which one is called upon first in understanding a sentence? And what
is the process that is occurring as one parses? Is it a stochastic process
similar to a Markov chain? None of these questions have been fully resolved,
but the evidence seems to be accumulating to suggest that syntax comes
before semantics (cf., Chomsky’s sentence).
One strategy proposed by Lyn Frazier, however, was called “late
closure”: “When possible, attach incoming material into the clause currently
being parsed” (Frazier, 1979:20). (It could be purely coincidental, but here
again a term used in graphic communication is also turning up in linguistics.)
Various other approaches to parsing were also proposed, and a good amount
of ink was spilled trying to resolve why some sentences, such as garden-path
165
and other ambiguous or confusing sentences, would cause people to come to a
wrong closure.
One possible answer to explain how parsing worked was to use the
“canonical sentoid strategy” or CSS, which posited the idea that a listener or
reader was actually using the underlying framework in English that
determines the order of sentence parts. Each language has its own pattern,
which is on a very abstract level. The canonical sentence for English is
subject-verb-object or SVO. The use of this strategy was ultimately put aside
by Frazier and others because they found that “To reconstruct the base
structure underlying a sentence, it is necessary to know not only which
sentoids occur in that structure but also how they are arranged relative to
one another” (Fodor, Bever, Garrett,1974:348). The strategy was, of course,
being applied to speech rather than writing. Because they were concerned
with speech, there certainly could not be any predictability in what types of
structures might occur and in what order so using the CSS was not a
confirmable strategy in the context of speech.
However, if we apply the CSS to writing, we may find the purpose for
some dynamic sentence punctuation. In contrast to speech, we already know
in writing what structures are there and in what order. All that is needed is
to use some system for showing the reader the relationship of the parts so
that the written language can be processed appropriately. In fact, it is the
abstract canonical sentence (SVO) that is actually being used as the guide for
166
the punctuation of sentences that are other than simple. To accomplish this,
only one punctuation feature is used: the comma.
The comma does not function in a positive way, however. It functions
in a negative way by showing which oncoming units should not be parsed
with 1) the canonical sentence or 2) some other immediately preceding
structure. In doing so, it is preventing closure. It is preventing the oncoming
or next-to-the-right structure from modifying the existing one to the left. The
comma is an “antimodification device,” an AMD. The same modification
prevention is also performed with recursion, with a series of words or
contiguous phrases of the same type: red, white, and blue; in the morning,
behind the garage, in the new car.
The comma, then, has two functions: 1) to prevent words or phrases
from directly modifying the canonical sentence (CSS); 2) to separate recursive
words, phrases, and sentences (the sentences when they are linked with a
coordinator).
If your sentences become too elaborate and you are reluctant to use
punctuation, your more complex sentence can easily become a nightmare:
The cat I’m referring to the neighbor’s cat which comes around
our house frequently begging for food something I don’t want to
give it because it might lose its interest in our neighbor and start
thinking it lives at our house is black and white.
167
What is needed is show the reader where the basic sentence is, and that can
be accomplished easily by using commas:
The cat I’m referring to, the neighbor’s cat, which comes
around our house frequently begging for food, something I don’t
want to give it because it might lose its interest in our neighbor
and start thinking it lives at our house, is black and white.
The punctuation makes the canonical sentence apparent by preventing
closure with structures that are not a part of it. This is not being
accomplished by using speech as a referent, by using logic or rhetoric as a
guide, or by using grammar in any ordinary way. The sentence now is visible.
The digressions in the middle of the sentence above are of several types:
apposition, relative clause, apposition, subordinate clause with coordination.
Each unit can be prevented from modifying the next by using a comma,
letting the reader know that the sentence is not ready to be resolved. When
the verb “is” occurs, the reader knows the sentence is going to be completed.
It is the word that the phrase The cat I’m referring to was looking toward
for closure and completion of the canonical sentence.
The parts of the cat sentence above could be arranged in a number of
ways, but the simple sentence could still be preserved using only commas.
Thus, the goal of dynamic punctuation is to preserve the integrity of the
canonical sentence and make it visible, and the comma is the chief
punctuation feature for doing this.
168
This can be observed in other sentences that we might use as
examples, regardless of what genre or source. Indeed, if we use this process to
reveal the canonical sentences in writing samples, we might also discover
places where good punctuation procedures have not been followed. Here is a
sample borrowed from Knobler that works well:
When it becomes necessary to communicate these sounds beyond
the limits of hearing, written symbols, a combination of lines
or marks, must be used; and these, too, require a common
understanding on the part of the communicators.
(Knobler, 1966: 24.)
Here is one that is questionable:
During this same period of time a minority of artists began
to experiment with new methods of representation.
(Knobler, 1966: 41.)
Here the comma has been omitted from an introductory phrase, something
which may cause a problem for some readers. (Indeed, a study of this writer’s
punctuation behavior would likely reveal that he is somewhat inconsistent in
the punctuation of introductory phrases.)
It should be made clear that each comma is functioning
independently, even though they sometimes appear to be working in pairs,
such as when a single apposition follows a noun phrase:
Jack, our neighbor, is very handy.
169
It is true that the two noun phrases could be interchangeable, but the
sentence began with the noun phrase “Jack,” and it is that one that is seeking
closure with the verb “is.” This is one of the peculiarities of the system that
may require some investigation. Some discussions of punctuation talk of
“paired commas,” but more likely each comma is dealing with “closure”
separately.
4.24 The Role of Interpolations
While the comma can be used to help place a word or phrase almost
anywhere in a sentence, interpolations can only be added after a sentence has
gotten under way. Interpolations are intrusions in a sentence but are
intended to add some kind of useful information to the sentence, usually in
proximity to something that has just been stated. The dash, parenthesis,
bracket, and quotation fulfill this function. They are visibly part of the
graphic sentence but are isolated from it by their punctuation. They do not
become part of the syntax of the canonical sentence. The ellipsis is used to
show omitted language where it is embedded in a quotation. It is an
interpolation inside an interpolation. Brackets can also be used as an
interpolation inside an interpolation ( [ ] ). Quotation can be done inside
parenthesis or brackets as well. Also, some portion of a quotation can
incorporated into a sentence of the author.
170
4.25 The Role of Adjuncts
The semicolon and colon almost always appear near the end of a sentence
because they are usually used to add some information to the basic sentence
in a qualified way. The special symbol “permits” them to make a semantic
modification to the sentence or partial. The semicolon has a dual role. It also
serves as a replacement for a comma between the units of a recursive
structure that is complex or as a replacement for “comma + coordinator”
between two sentences.
4.26 The Role of terminal punctuation
Terminal punctuation and initial capitals have a single semantic goal: to
show the end and beginning of an orthographic sentence.
4.27 The Absorbtion Rule
When two kinds of structures (ex., a dash and a comma) could possibly end
up in the same location, the one that is more likely to have been there to
preserve the canonical sentence would take precedence.
* a) The house, which is dark—because nobody is home—should be shut.
b) The house, which is dark—because nobody is home, should be shut.
Normally, we see dashes in pairs, but in a) the right dash overlaps with the
location where a comma would normally be if the dash had not been used. In
this kind of situation the absorbtion rule comes into play and the comma in b)
171
must replace the right dash. Otherwise, the canonical sentence would be
sundered. The dash belongs to the relative clause digression, not the
sentence.
4.28 Summary
Punctuation is a symbol system in a graphic domain. It needs to be analyzed
in accordance with gestalt and other visual and graphic standards in
addition to how it functions with linguistic categories. But the visual model or
schema that it uses is not just ordinary grammar but instead the more
abstract canonical sentence (SVO).
The action involved in all punctuation is modification or its
prevention. The written sentence is an orthogonal projection and has two
dimensions, static and dynamic. Static (vertical) punctuation (capitals,
underlining, italics, bold, hyphens, apostrophes, etc.) semantically alters
graphemes phonologically and morphologically. Terminal punctuation is also
static and semantic. The goal of dynamic (horizontal) punctuation is to
preserve the integrity of the canonical (SVO) sentence and make it visible by
using commas to prevent inappropriate modification. Parentheses, brackets,
quotations, and dashes are interpolations and not part of the sentence itself.
The semicolon and the colon are sentence adjuncts. Ellipsis is an
intraquotational punctuation feature. The absorbtion rule handles right
closure punctuation conflicts. Punctuation involves markedness. Punctuation
172
is grammatical only in the sense that punctuation features must be inserted in
the proper places once their relation to the canonical sentence has been
identified and the canonical sentence is protected and made visible.
173
Chapter 5
An In-depth look at some Punctuation Features; some Anomalies;
the Puncteme; and some Research and Teaching Ideas
5.1 Introduction: Speaking and Writing
As long as people continue to perceive writing as closely related to speech,
they will have trouble with punctuation. Writing is graphic. It must be seen.
It takes after its classification: orthographic. Speech is like a massive
symphony orchestra whose music can be broadcast to the entire world via
satellite. Writing is barely a thin octet playing to an audience of one. Speech
has a huge advantage over writing with nearly unlimited kinds of nonverbal
communication to assist it. A wink and a groan can mean much more than a
five hundred word essay. A few decades ago, there was a considerable
interest in comparing speech and writing, but there has been little interest in
doing so recently.
When teaching the alphabet and writing, it is necessary to show the
similarity to speech, but the differences should also be taught, otherwise
students may get the impression that they are only mirror images of one
another. Alphabetic languages have advantages; they can invent new words
easily, something that syllabaries and logographic writing systems have more
trouble doing. But very likely someone used to a written language like
174
Chinese would find an alphabetic language just as confusing as someone
writing in English would with Chinese at first encounter.
5.2 The Comma as a Tool
The comma is the only feature of punctuation doing all the work in the
horizontal or dynamic axis. It is the greatest challenge for many writers and
has the greatest number of “laws” or “rules” in handbooks. It has also been
identified as the most common punctuation feature by Charles F. Meyer, who
found that 47% of the punctuation in the corpus he studied was commas, with
periods taking second place at 45% (Meyer, 1983:18). No other feature comes
even close, so it would benefit new writers to learn the two uses of the comma
first. It also is the cause for a large number of writing errors. An article by
Connors and Lunsford on student writing errors revealed that comma errors
were one of the most common (Connors and Lunsford, 1988).
To understand the comma’s uses it is necessary to conceive of writing
as graphic. The comma is the only feature of punctuation actually used in the
syntax in the horizontal axis. It has two uses:
a) to protect the canonical sentence by making sure that there
are no inappropriate modifications;
b) to reveal recursions.
In effect, both are preventing modification, so we could possibly talk about a
single task for the comma. But there is a difference in the places where you
find them. The comma protecting the canonical sentence gets its direction
175
from the invisible form and involves parsing, while the recursive comma gets
its direction from visible and adjacent repetition in the axis.:
c) The long, dark, curvy road was dusty.
d) I drove carefully, slowly, intrepidly.
Omitting the commas would allow each adjective or adverb to modify the next
one (which could be an option if the writer intended that meaning).
Coming to grips with the idea that modification is something going on
all the time in a sentence could have some advantages. For example, two
structures which are exactly the same can mean two different things in a
sentence, depending upon whether you add a comma or two to prevent
modification. A relative clause (and similar structures) can be used to modify
a noun e) or just add some interesting but not necessary information f).
e) The man who flew upside down was crazy.
f) The man, who flew upside down, was crazy.
In e), the relative clause is modifying the noun “man.” In f), it is just some
information that is interesting but not crucial to identifying the man and
could be left out. If e) were called “modifying” and f) were called
“nonmodifying” (rather than “restrictive” and “nonrestrictive”), our
understanding of both grammar and punctuation might be improved.
Punctuation, like other linguistic categories, is also demonstrating
contrastive distribution of a symbol in the same context. Since the comma is
176
the only feature in the syntax (the horizontal axis), it is usually
comparatively easy to tell the difference between the two uses.
The comma in a) is a model which can be used in the various kinds of
sentences. Those parts of an orthographic sentence, which may or may not
resemble a canonical sentence, also use commas in the same structural
locations where they might be used in a canonical sentence. A partial
sentence can also contain commas in the same way, and its difference is that
it does not have a junction between a subject and a verb.
When there are two a) commas involved in a potential modification
location, they are sometimes described as being “paired.” But they are really
working individually. As the sentence is parsed, the first comma is noting the
potentially inappropriate modification and signals it. The attention is then
directed to the modification itself, which temporarily becomes the focus until
it comes to the point where it no longer is grammatically appropriate for
what is coming next. Then another comma is required, not because it is
related to the previous comma, but because another inappropriate
modification may now be about to take place in that location. The sentence is
always parsing forward, and each comma gives permission to delay the
semantic closure of what just came before to what is coming next. (If the
second comma occurs at the end of a sentence, the absorbtion rule comes into
play, and terminal punctuation is used instead.)
177
5.3 Recursion and the Comma
The comma operates in a “series” as an indicator of recursion. We are often
told that a series is at least three things, but it really need be only two.
Recursion is any two or more units in a row that are grammatically
equivalent. This is really going on all of the time in the language—for
example, when you put several prepositional phrases in a row (“in the
morning at the hotel in New York”), but sometimes there is the need to
isolate each unit by using commas and creating a list-like arrangement which
is more noticeable. The goal again is to prevent one unit from attempting to
modify the next one.
The commas are the feature that is doing the work in the series, but
often there is an “and” between the last two items. Style then may come into
play, and some writers choose to eliminate the final comma, creating the
potential for ambiguity. In fact, because of the recursion and the commas, the
“and” is the feature that is not needed and can be eliminated without altering
the meaning in the series. The formal name for this is “asyndeton” or “no
coordinator.”
The comma is also used to combine two (recursive) sentences, but this
requires an additional feature to make sure the reader notices it is a different
level of recursion. Some function word is needed in addition to the comma: S,
and S; S, but S; S, yet S; S, or S. If the function word is omitted, however,
the sentences can be combined with a semicolon: S; S. In either case the
178
recursion is noted. One comma error involves putting a comma after a
sentence, followed by a coordinator and a phrase (S, and phrase), often based
on some idea of a pause, and deceiving the reader into believing another
sentence is imminent.
5.4 Interpolations
Robert H. W. Waller defines an interpolation as “the insertion or
juxtaposition of a short segment into a longer one in such a way that the
continuity of the sentence, paragraph, page, chapter, or book is not
destroyed” (Waller, 1980:248). In graphology, you can account for variables by
making one of them stable and treating the other as an interpolation (inter –
between, among; pol – an axis) ( Mandell, 1974:99-101). The invisible
canonical sentence (SVO) in this case is the stable variable, the horizontal
axis. Additions to its own parts are indicated with the comma. But additions
to the sentence at large can also be made using interpolations, which are a
linguistic set of features. There is no clear hierarchy among them, but some
seem to have a greater distance from the basic sentence than others, but a
subset may function as a hierarchy if necessary. (Other punctuation features
can also be found inside interpolations of course.)
The dash < ─ > probably has a closer relationship to the basic sentence
than the other interpolations do, because it is “empty” and functions like
empty spaces, yet it interacts with other syntactic punctuation in the
179
horizontal axis, such as the comma or terminal punctuation, where the right
dash can be absorbed. It is something of a hybrid. All interpolations perform
like sotto voce, but the dash seems the least in that regard, more often
indicating a less than important reason for departing and a rapid
interruption and return. All of the interpolations may show some stylistic
effect, depending upon how they are used. (However, stylistic effects in the
basic sentence are more likely to be caused by word choice, word order, and
larger features of the writing rather than punctuation, except for poetry,
perhaps.)
The small subset involving the parenthesis < ( ) >, brackets < [ ] > , and
braces < { } > does form something of a hierarchy, each representing more
distance from the basic sentence, and each can be inserted in the previous
one, but that much embedding might be extraordinary < ( [ { } ] )> and
difficult to follow. Parentheses often provide some useful additional idea,
whereas brackets tend to include technical or editorial information. Braces
are seldom used.
There is also the lurking problem of semantic meaning involving
interpolations. The comma deals with semantic alteration in the meaning of
the basic sentence, while interpolations seem to have more of a structural
approach in addition since they are interrupting the basic sentence and
announcing that they are doing so to provide some kind of modification on a
different level—factual, technical, or some kind of personal observation, for
180
example [cf., Montague grammar, where the context plays a role in the
meaning (Lappin, 2001:375)]. Still, these interruptions do alter the meaning
of the basic sentence.
There is also the possibility that some interpolations are involved in
metalanguage, particularly if they are used to define or clarify some language
in the basic sentence. The force of punctuation can be so great, in fact, that at
least one scholar has claimed of some that it “was a kind of commentary of
the text” (Saenger and Heinlen, 1991:255).
Interpolations also have constraints. The ones listed so far (dash,
parenthesis, brackets, braces) almost invariably follow the part of a text that
they relate to or may show up at the end of a sentence. You can’t start a
sentence with an interpolation, a hyphen, or an ellipsis (at least not yet in
ordinary prose), but you can with a quotation or an apostrophe. On the other
hand, interpolations aren’t likely to have any special meaning outside the
sentence in which they exist; they are context bound. Also, they may be
subject to editorial practices, like all the rest of punctuation.
5.5 Parenthesis and Quotation
Over the years, some scholars have become fascinated by certain features of
punctuation. Two that have garnered particular interest are parenthesis and
quotations. Book length works and many articles have been written on both
features.
181
John Lennard, in his 324 page book, But I Digress: The Exploitation of
Parentheses in English Printed Verse, makes it clear that “parentheses” can
mean at least two different things:
‘parentheses’, rhetorical figures; and ‘parentheses’, the marks of
parenthesis sometimes called “round brackets”: ( ).
(Lennard, 1991:1.)
Lennard’s interest was in the “round brackets,” also known as “lunula (-ae)”
or “little moons,” which he borrowed from Erasmus, he claimed (Lennard,
1991:1). The symbols themselves, according to Paul Saenger, were invented
by Humanist scribes in the fifteenth century and spread throughout Europe
quickly (Saenger, 1982:410-411). The scribes were actually just giving a form
to the rhetorical figure that had been around already, a Greek expression
meaning something like “alongside.” Apparently, it didn’t take long for the
little moons to become popular. One study cited by Lennard revealed that
Shakespeare’s play scripts were littered with them.
But Lennard’s goal was to trace the use of parentheses over the
centuries but only in verse. One of the conclusions he reached was that
opinions on the use of parentheses differed strangely. Grammarians felt they
were “additional, irrelevant, extraneous, subordinate, or damaging to the
clarity of an argument,” while what he observed himself revealed that “they
were often original, relevant, central, emphatic, or indicative of the crux of an
argument” (Lennard, 1991: 242). Lennard discovered that little moons have
182
remained popular, something that anyone who has read poetry by e e
cummings can attest to.
Parenthesis has also remained popular in prose as well. Daniel P.
Deneau did a study of Claude Simon’s 1960 novel The Flanders Road, where
he found “550 sets of parentheses which enclose approximately 25% of the
total text” (Deneau, 2003:553). Deneau discovered that Simon’s prose was
certainly unusual but his use of parentheses was extraordinary, for what was
in them was as important as the rest of the text, despite a popular belief that
parenthetical material can easily be omitted from a text without losing
anything. Simon’s prose is indeed unusual, with sentences running for two or
more pages and parentheses the length of an average paragraph at times.
(Trying to provide an example here could well exceed the legal limit for
citations.) Apparently, the little moons are still shining brightly in prose as
well.
Of all the symbols, quotation seems to have attracted the most interest
by scholars. Like other punctuation features, quotation symbols have more
than one function. But the scholarly interest in quotation, largely, is not in
the symbol itself but in how the two types, 1) the words of others or 2) words
just getting some attention, can fit into the linguistics of a sentence.
Currently, the symbol has only those two main uses: 1) to indicate a
direct quotation from some person or place other than the author of the
present document; 2) to call attention to some language that may require a
183
second thought for any one of a number of reasons (one type is referred to as
“scare quotes”). Once referred to as “inverted commas,” the symbol itself is in
effect similar to italics, bold, underlining or font change in its graphical use,
except that its attention-getting function is meaningful and specific rather
than like the more general and stylistic uses of the other features. In both
uses, the reader is being advised to notice the language itself, to pay
attention not only to the “what” of the language but also to “how” it is said
and “who” said it. In a sense, the language in question is more “objectified,”
rather than merely to be parsed and understood as just another part of the
author’s argument. Or, as Townsend and Bever put it, “when a word
sequence is organized into a sentence, it takes on a special psychological
status” (Townsend and Bever, 2001:45). This certainly would amount to a
semantic change when quotation symbols are used.
Many articles and some valuable books have been devoted to
quotation, but mostly to its linguistic or literary functions rather than its
graphical use. According to Colette Moore, author of Quoting Speech in Early
English, today’s symbols were a late arrival, sometime after the start of
printing. The early parentheses were also used to indicate language other
than an author’s, but the whole practice was somewhat haphazard. Moore
claims that the current symbols were already in some books in the sixteenth
century, however (Moore, 2011:76). In addition to Moore’s book, Marjorie
Garber’s scholarly work titled simply Quotation Marks looks into the “uses
184
and abuses” of quotation in a variety of contexts. For someone interested
mostly in the linguistics of quotation rather than just the uses of the symbol,
a recent work is Understanding Quotation, edited by Elke Brendel, Jörg
Meibauer, and Markus Steinbach (2011).
5.6 Diacritics. Diacritics share the vertical or static axis with capitals and
font alterations. They may be the most debated punctuation, and the attempt
to avoid them may be responsible for some of the odd spellings in English.
That is, to avoid having to use diacritics to provide more accurate graphemes
in English, printers probably encouraged using other graphemes instead.
Diacritics indeed are of two types, one involving adding an additional
grapheme and one which modifies a grapheme (Aronoff, 1994:74). For
example, the silent <e> in “write” and the doubled <tt> in “written” could
likely to be the result of printers not wanting to adopt diacritics to modify an
individual grapheme as is done wholesale in languages like Czech (Aronoff,
1994:77).
Even some linguists have trouble acknowledging the existence of
diacritics in English. In the punctuation section of The Cambridge Grammar
of the English Language, everything on the punctuation list we have been
talking about in this study is included except one, about which they say:
“Words may also contain various non-segmental marks, diacritics, but we do
not regard these as falling in the domain of punctuation” and are “simply a
185
matter of word-spelling” (although they are unsure about the diaresis <ӓ>)
(Nunberg, Briscoe, and Huddleston, 2002:1724-1725). But they apparently
agree with the idea that the twenty-six minuscule letters are the basic form
of the graphemes in English because they state the following:
We will therefore regard punctuation as covering the use not
only of punctuation marks but also of such non-segmental
features as italics, capital letters, bold face, and small capitals.
Ordinary lower-case roman represents the default form, and
these non-segmental features can be regarded as modifications
of the default form.
(Nunberg, Briscoe, and Huddleston, 2002:1724.)
Why they put diacritics aside is hard to understand, but they do support the
basic graphemes and believe in the concept of modification.
Gabriel Altman had no problem with diacritics as punctuation and
described the situation exactly: “One can count also the diacritical marks as
grapheme modifications . . . .” (Altman, 2008:154). Alan S. Kaye, disturbed by
an article claiming that English used no diacritics, studied Time magazine for
a few weeks and found these: cedila, accent acute, accent grave, circumflex,
diaeresis, macron, tilde, and spiritus lenis (Kaye, 1988:11).
Dennis Kurzon, in “A brief note on diacritics,” discussed some of the
arguments about them. Some believe that for something to be considered a
diacritic in a language it has to be a regular feature, but that would be
186
making a decision based simply on how often it was used (Kurzon, 2008:90).
Someone driving past a Café every day might argue with that assumption.
(There is at least one business in Milwaukee that uses an umlaut in its name,
which is on the façade of the building where it is located.)
5.7 The Apostrophe
To Elizabeth S. Sklar, the apostrophe is :
. . . the stepchild of English orthography. It is neither fish nor
fowl, typographer's convenience, nor true punctuation.
Ordinarily well-behaved, the apostrophe is usually seen but not
heard, a device for the eye rather than for the ear; yet unlike
any other orthographic symbol, be it "point" or printer's mark,
the apostrophe, under certain phonological conditions, is
pronounced as a full-fledged phoneme. The possessive
apostrophe is a grammatical anomaly, a vestigial case
marker—appropriately shaped like the human appendix—in a
noun system that has otherwise dispensed with cases.
(Sklar, 1976:175.)
It seems apparent that Sklar does not like apostrophes, and she also
cataloged in her article a number of street signs and other places that
incorrectly left them out.
187
J. C. Wells, in an article on diacritics, suggested that the apostrophe
can sometimes be considered one: “. . . from some points of view its role is
considered similar to that of the diacritic. It is true that in many languages it
is used to indicate the omission of a letter: this is how it is used in
English . . .” (Wells, 2000:253).
Its physical appearance, perhaps, has caused some to call the
apostrophe a clitic, something which is “less than a word but more than an
affix,” for which Trask provides “she’ll do it” and the possessive “ ‘s “ as
examples (Trask, 1997:44). In another work, Trask identifies the apostrophe
as a grapheme, along with all the other features of punctuation (including
space) (Trask, 1999:114).
To Daniel Bunčić, the “apostrophe does not mark the omission of
letters, as traditionally assumed, but indicates important morpheme
boundaries wherever this is necessary for certain reasons” (Bunčić, 2004:
185).
Edwin Battistella did a study of the apostrophe as related in a way to
Peirce’s sinsigns and legisigns, but made no determination as to how they fit
into the whole process of punctuation or how this idea relates to how other
punctuation features function (Battistella, 1999).
There are a number of opinions, then, as to what an apostrophe is:
diacritic, clitic, modifier, boundary marker, and sinsigns and legisigns. But
Quirk et al. did manage to find something to make the apostrophe more
188
linguistic when they said that “There are occasional examples where the
genitive acts as a modifier rather than as a determinative,” and offered words
like “women’s,” “ship’s,” and “farmer’s” as examples of a “descriptive genitive”
(Quirk et al., 1985:327).
Indeed, like other features, the apostrophe is a contrasting pair, with
two different uses in the same context, one for an omitted grapheme in a
word and another to show possession. Both occur in the static (vertical) axis
in the sentence, and the two uses almost seem like opposites: the possessive
apostrophe shows a concept added and the other use shows a concept left out.
The apostrophe can be used nearly anywhere to show an omission, even the
beginning of a word ( ‘T was) or sentence, but the possessive is used only at
the end of a word. The well-recorded problem with the possessive apostrophe
has to do with its use with words ending with “s” (in which case 1] just an
apostrophe is added and no additional “s” or 2] another “s” is added after the
apostrophe) and words which can’t logically possess anything, such as “my
car’s tires are flat,” which disturbs some readers. These could easily be called
“allopuncts” to confirm their legitimate linguistic character, but writers are
definitely puzzled with how to use them.
Sklar’s advice, you can imagine, is to get rid of the possessive
apostrophe:
While the apostrophe is useful in the world of letters, a
reminder, perhaps, of some more formal and elegant stage of our
189
language, it is, in the end, an antique. . . . When it ceases to be
a convenience, we have no option but to allow the apostrophe to
join the flatiron, the washboard, and the footwarmer as a relic of
times past.
(Sklar, 1976:183.)
5.8 The Hyphen, Ellipsis, and the Virgule
Like other punctuation features, the hyphen has two contrasting uses: 1) to
show a connection between morphemes and words and 2) to show separation
of morphemes. It operates on the vertical (static) axis. The hyphen at the end
of a line to indicate a broken word is a problem, however, since it can be
confused with the other use of the hyphen. This could be solved by modifying
the hyphen which is breaking a word at the end of a line, perhaps tilting it a
bit, to show a difference that distinguishes it from the connecting type of
hyphen.
Ellipsis may be the only feature of punctuation that seems to have
nothing to contrast with, unless we contrast it to space, which is in a different
axis. Its use is very specialized: to show purposely omitted language in a
borrowed quotation. So the ellipsis is nearly always visible inside a line of
text that is already framed as a quotation. Fiction generally uses ellipsis
more generously, but there the dialog is not in always in quotation marks.
One curious thing about the ellipsis is the way it is treated in some European
writings. It is sometimes enclosed in brackets: [ . . . ]. Apparently, some
190
printers do not believe that spaces are sufficient punctuation. The brackets
are clearly redundant.
The virgule, also known as traits d’ union, turned up in Europe “in the
late tenth and eleventh centuries,” according to Paul Saenger (Saenger, 1997:
66). It was also known by some other names and used in a number of ways
over the centuries. Today it usually indicates a choice or equivalency between
two ideas (perhaps opposites again). Bringhurst also claims that in the past it
was used as an alternative to the comma or parenthesis (Bringhurst, 2005:
81-82).
5.9 Capitals and Terminal Punctuation
The capital at the beginning of a sentence creates a problem similar to the
hyphen at the end of a line. The reader can’t tell if it is just a sentence initial
capital or also a proper noun, its contrasting use.
*Frank is the politician who will fully explain a new law. But they are rare.
frank is the politician who will fully explain a new law but they are rare.
Frank is the politician who will fully explain a new law. He always does.
The solution, of course, would be to eliminate initial capitals, but that might
take some energy to accomplish. If we could accomplish it, we might also be
able to eliminate the redundant period at end of the sentence as well and only
add terminal punctuation for questions and exclamations and initial capitals
only for proper nouns.
Josef Vachek also believes that capitals are a problem:
191
. . . the use of capitals to open a sentence is, in fact, redundant,
because the beginning of the new sentence is sufficiently
signaled by the presence of the full stop at the end of the
preceding sentence (or by the absence of any preceding context).
(Vachek, 1989:49.)
Vachek does lean a bit toward keeping capitals for the same reason—
redundancy, which he believed might in some way assist reading. For
example:
more than likely, the capitals and periods are barriers to faster
reading no one really needs them a reader can easily adjust to a text with no
obtrusive barriers, which I’m sure reading specialists would be able to
research and test, as they have with saccades, the units that a person scans
when reading
Downing and Leong, in Psychology of Reading, do provide some
information about the structural appearance of a text it does affect the visual
attention of a reader, they found one would suspect that the fewer visual
distractions there are the better!
(Downing and Leong, 1982:145.)
Herbert Bayer also believed that typographic elaborations should be
removed from texts to make them more readable. He omitted capitals and
serifs. It seems to work. Here is a sample of his ideas:
typography is a service art, not a fine art, however pure and
elemental the discipline may be.
192
the graphic designer today seems to feel that the typographic
means at his disposal have been exhausted.
accelerated by the speed of our time, a wish for new excitement is
in the air.
“new styles” are hopefully expected to appear.
nothing is more constructive than to look the facts in the face.
what are they?
the fact that nothing new has developed in recent decades?
the boredom of the dead end without signs for a renewal?
or is it the realization that a forced change in search of a “new
style” can only bring superficial gain?
it seems appropriate at this point to recall the essence of
statements made by progressive typographers of the 1920s:
previously used largely as a medium for making language visible,
typographic material was discovered to have distinctive optical properties
of its own, pointing toward specifically typographic expression.
typographers envisioned possibilities of deeper visual experiences
from a new exploitation of the typographic material itself.
they called for clarity, conciseness, precision; for more
articulation, contrast, tension in the color and black-and-white values of
the typographic page.
(Bayer, 2009:44.)
[Note: The font used in this borrowed example of Bayer’s text still had serifs. There also
is evidence that some small kinds of serifs can help in reading.] Bayer omitted capitals,
193
but he kept terminal punctuation. He also arranged sentences and parts of sentences in
patterns that might make them easier to read—once you got used to the arrangement.
Getting rid of capitals would also allow writers to start a sentence with a numeral,
something which is not good style presently. The ideal layout and punctuation for a page
of text may not have been achieved yet. There may be some room for improvement.
The contrasting use of the period is to show abbreviations. The exclamation and
question features have only a single use each, to modify the entire sentence.
5.10 The Adjuncts: The Semicolon and the Colon
By the far the most mysterious symbols in a text, the semicolon and the
colon, get more attention on average than the comma. Just the mention of
them in some company elicits responses that are extreme. Because they are
positioned in a context controlled mainly by modification involving other
punctuation, they appear to be out of place, but this is because they operate
on a different level or for somewhat different purposes.
The other symbols operate on local structural levels, separating
clauses, phrases, and words in the horizontal axis. Interpolations have
permission to interrupt nearly anywhere but at the beginning of a sentence.
But the semicolon and colon seem to operate at a quasi-textual level. They
are something like textual shortcuts imbedded in a grammatical string. The
result is that they are relegated to only certain places in a sentence and are
not found between words or ordinary phrases at random. They operate under
constraints. They don’t usually appear until after there is a complete
194
sentence and only in special places, suggesting that they have a subordinate
role in the punctuation system.
Like other symbols, the semicolon has two contrastive roles. It is
usually found in only two principle locations: between two complete sentences
or as an upgrade from a comma in a somewhat complex recursive situation.
In one use, it is a space-saving feature which makes it clear that more words
could have been spent on the language in question but that the symbol is
making it unnecessary. In that sense, the semicolon is also a modifier, but on
a seemingly more graphical level. The other use seems more mechanical: just
a comma upgrade, but it is also used in a complex series to replace a comma.
Paul Bruthiaux believes that the semicolon arrived in English around
1560 and “gained considerable popularity among literary figures” by the
middle of the seventeenth century (Bruthiaux, 1995:3). But expressive as it
was to some early adopters, the semicolon lost ground in the last five
hundred years, peaking in the eighteenth century and bottoming in the
twentieth century (Bruthiaux, 1995:7). He did a study of a corpus of 30,000
words assembled from works devoted to grammar, language, and linguistics,
places where he might find relevant information (Bruthiaux, 1995:5). He
attributed the decreasing interest in the semicolon as resulting from the shift
in the role of punctuation from prosody to syntax (Bruthiaux, 1995:9).
Eric Partridge, in You have a point there, has investigated most
punctuation features and provided examples of their use. (Unfortunately, he
195
didn’t document his sources.) He provided eleven examples of the how he
believed the semicolon was being used. Three of them are somewhat
questionable (they could probably be called purely “technical”), and the others
could probably be reduced in the way they are used to two core ideas: 1)
something series-like and 2) a semicolon between two sentences, sometimes
with a connective (Partridge, 1953:44-49).
Geoffrey Nunberg, in a talk on the semicolon at a Modern Language
Association meeting, aired similar problems with the semicolon (Nunberg,
1998). Is it paratactic (joins text clauses)? Is it hypotactic (introduces clausal
adjuncts)? Is it just a subordinator? It is under constraints: you can’t use two
in a sentence (other than the series, apparently). Nunberg’s talk raised many
questions but provided no easy solutions for a symbol that is supposedly
useful but whose application is puzzling. Teachers as a rule don’t seem to find
good ways to teach it.
Not everyone may agree. Angela Petit might not. She teaches the
semicolon as a symbol of style and rhetoric to her students using Martin
Luther King Jr.’s “Letter from Birmingham Jail,” which probably includes
more than the average number of semicolons for a text its size (Petit does say
that there are different versions of it in print, however [Petit, 2003:69]).
A look at King’s letter, in this case the August 1963 version in the
Atlantic Monthly, reveals that King’s style actually fits the more usual
patterns quite closely, at the most showing something of a pulpit style in the
196
way he used recursion by echoing a speech technique (King, 1963:78-88).
Seventeen of the twenty-nine semicolons in his letter are used to separate
recursive clauses preceded by “when,” “who,” “let,” or “if.” Another nine are
simply separating two sentences (S; S), and two are separating sentences
that include a coordinator (S; and/but S). Petit’s use of this letter is
admirable, certainly. Showing the use of a feature of punctuation is a good
way to get students to understand it, a way that could be called “the
reputable writer’s approach.”
But this brings up a question: Where is the style? Is it in the language
or is it in the punctuation? Which is the controlling factor? Does the language
create the need for certain punctuation, or does the punctuation create the
need for certain language? This is certainly a question that needs some
research along with similar situations involving punctuation. What can be
said about the intersection of meaning and form in the way punctuation is
used? And in what way, if any, does the semicolon get involved in
modification?
Neither the semicolon nor the colon can be found at the beginning of a
sentence. Both require some introduction by way of a sentence. Although they
are often dealt with together, as we are doing here, it could be that they are
independent of one another. The semicolon, we may have established, has
two principles roles, a characterization that fits with other features. Is it
197
possible, perhaps, that the colon is also an independent feature unrelated to
the semicolon and has more than one role?
The colon is truly an orphan where scholarship is concerned. Studies
that are available are largely devoted to its history or its use as a predictor of
scholarly acumen with its use in the titles of research work. J. R. Dillon, in
the 1980s, studied seventy-two titles of scholarly work in thirty journals over
a century. His conclusion was that the colon showed only minor surges in
activity until 1950, when the use of the colon in titles showed an exponential
leap in use just before 1980 (Dillon, 1982:96). His interest in the phenomenon
spawned research on the subject by others.
To find a coherent assessment of the use of the colon is difficult,
suggesting that, although it has several “technical” uses (such as introducing
lists of items or preparing the reader to focus on something), a sole or best
use of the colon otherwise seems nonexistent. Some uses fall into general
categories like problem/solution, question/answer, statement/explanation,
generalization/details, and similar dualities. On the other hand, perhaps this
is the second role of the colon after all: to visibly reveal arguments that have
two parts, one on each side of the symbol. But there is no easy linguistic
description for these two roles, the technical uses and the more meaningful
pairing; that may have to wait for more research. Maybe the colon will turn
out to be a Rosetta Stone of punctuation.
198
Adjuncts, then, seem to operate on a more organizational level,
perhaps, as a way to overcome some of the shortcomings and otherwise
simple punctuation in the syntax, especially that of the comma, which has
other duties to keep it busy.
5.11 Graphemes and Punctemes
When we visit a doctor for a complaint, we hope to leave with a name for
whatever ails us. There is a perverse satisfaction in this. We like to know
what to call things that we need to talk about with others. But there are
some drawbacks when a name for something does not quite fit the facts. The
term “punctuation mark” is this kind of problem. The use of it actually
prohibits a person from looking for, as punctuation, anything that doesn’t
look like a “mark.” I have carefully avoided the term throughout this study
and used it only when it was in a quotation. The problem is that it is
generally much easier to keep using a term we know already than to go
looking for a new and more accurate name for something, even if we suspect
the current term is not right. In addition, the term “punctuation mark” does
not fit with the other terminology of linguistics.
Some of the gestalt psychologists did experiments that involved
problem solving and discovered that, if a person already had some kind of
solution for a problem, they were reluctant to go looking for a new one.
Richard E. Mayer reported on some experiments which revealed that past
199
experience in an area proved to be a block he called “functional fixedness”
(Mayer 1995:17). The use of the term “punctuation mark” probably involves
this kind of mind set. It characterizes punctuation only as independent
visible forms and completely rules out spaces and grapheme alterations.
But before we start looking for a better term, we might want to put the
blame for the existence of “punctuation mark” where it belongs. That may be
hard to determine. The expression is in grammars and related books at least
as far back as the early 1700s, and finding the ultimate source may be an
impossible task. Obviously, the term has outlived its questionable usefulness
and should be replaced with something that fits better with linguistic
discussions about punctuation.
As a replacement for “punctuation mark,” I propose that any feature
that we determine is involved in actually punctuating be called a “puncteme.”
Like all the other “-eme” units, punctemes must be individually identifiable
items that are part of a set and operate as a system. This would also create a
coherent alignment in the way punctuation items are described and
categorized. They are a limited class. They also have no independent
meaning. To complete the picture, we need to describe what punctemes do:
Punctemes are symbolic features which function in the English
writing system in one of two ways: they modify some units in the
writing system or they prevent the modification of other units.
200
The units that are modified directly tend to be phonological or morphological
or are in the static axis. The units for which modification is prevented are in
the dynamic axis. A few punctemes can modify an entire sentence or a partial
one, such as the terminal punctuation which modifies a sentence. Spaces
appear as modifiers in both axes and elsewhere.
Like their other “-emic” relatives, all punctemes have functional
relationships and make up a closed system of contrastive and minimally
distinctive units. Looking at it another way, you need to have a phoneme to
have a grapheme, but speech has no “–emes” that punctemes can be derived
from or represent in writing. They are graphic symbols created exclusively for
a visual written version of English. The study of punctemes, of course, is
“punctemics.” We can imagine Walter Ong calling them “tools we invented to
adjust the technological features of writing and make them more
immediately understandable.”
5.12 Punctuation and Research
The long discussion about the basis for punctuation has probably gotten in
the way of other kinds of research that could have been done on written
materials of the past in English. Perhaps a new way of looking at
punctuation will open some doors and provide some new techniques for
investigating older materials.
201
One approach might be to apply the canonical sentence schema to
older writings and, with a more standard view of the sentence, see just
exactly what earlier writers were up to in applying punctuation to their
writing. If they were using some grammar, for example, that might be
evident. If they were using some approach to prosody common in their time,
that also might be evident. If they were using punctuation merely as a
version of their own personal style, that might be more easily determined.
This approach could also be applied to a study of a period of English. There is
some evidence that English changed from an SOV language to an SVO
language in an early period. This might be traceable using a more definite
view of the sentence and any possible punctuation if enough documents were
available. Also, the history of the sentence itself might be better understood
with a schema approach.
Special kinds of punctuation could be understood better with a good
model to work with. Ian Robinson, in his article on Medieval Punctuation, the
Concept of the Sentence, and Reading, has suggested that the “singing” verses
of Middle English used punctuation and would be worth doing some research
on (Robinson, 1992:43).
The English syntax of the past millennium might be easier to
understand using a standard approach rather than a vague approach linked
to either grammar or prosody. Since writing is a sort of doppelgӓnger of the
202
speech system, some better understanding of writing might also tell us more
about the speech behavior of earlier periods.
The writing in the earlier periods in the history of English often
included punctemes that are no longer used. Some research on them might
shed some light on how and why they were used.
5.13 Manuscripts and Punctuation
In recent years, there has developed more interest in who really applied the
punctuation to many printed manuscripts. Rumors abound about the
involvement of printers and their aides in the punctuation of manuscripts. A
new approach to punctuation might help determine the person who actually
did the work, possibly by comparing manuscripts by different authors printed
by the same printer. Some manuscripts by older writers have been
repunctuated by editors, possibly removing some important features that
were intended to be there by the authors. This could be a rich avenue for
punctuation research.
5.14 The Forensic Study of Writing
Much has been written about who really wrote Shakespeare’s plays. Two
books that dealt with his punctuation are A. C. Partridge’s Orthography in
Shakespeare and Elizabethan Drama and Anthony Graham-White’s
Punctuation and Its Dramatic Value in Shakespearean Drama. Besides
203
looking at fingerprints and DNA, a writer’s punctuation could easily become
a factor in author identification. Don Foster’s forensic approach to
authorship, Elegy by W. S.: A Study in Attribution (a study of a poem that
could have been by Shakespeare) and Author Unknown: On the Trail of
Anonymous, might have benefited by using punctuation as well as other
features of language to determine authorship.
5.15 Punctuation for New Written Languages
New writing systems are being devised on occasion, and there certainly will
be the need for punctuation in them. Ettien N. Koffi’s article, Indigenizing
Punctuation Marks (1995), described the problem of deciding just how many
and which kind of punctemes to include in a newly written language. A better
understanding of how punctuation works might make that task easier.
Old languages that adopt new punctuation sometimes have trouble
adjusting. William C. Hannas, author of Asia’s Orthographic Dilemma, has
noted the problems that East Asians have had by ignoring the punctuation in
writing: “Although texts in the four major East Asian languages today are
punctuated, much of it seems to be done as an afterthought” (Hannas, 1997:
263). The problems multiply when electronic technology demands better
adherence to some kind of conventional punctuation. A better approach to
punctuation might help regularize a failing system.
204
The areas of study that are involved in electronic approaches to
language could also benefit from a new way of seeing punctuation. A number
of the dissertations noted in Chapter 3 having to do with computational
linguistics and natural language research made it clear that the older
approaches to punctuation did not solve their research problems, nor did
leaving some of the punctuation out of studies.
5.16 Punctuation and the Future of English
Although we do spend some time looking at the past of our language, we
probably seldom think about the future of it. It may be time for a diachronic
study of the punctuation of English with a view toward considering
improvements in the way punctuation is used and devising new punctemes to
deal with issues that seem to occur with some regularity. This does not mean
that we need a bureaucracy to assign punctuation standards, as Germany
has recently done, but perhaps some scholars might look at possible
improvements and report their findings. A good number of people have
already recommended new punctemes to enhance the system we already
have. New ideas seem to show up every day. I. A. Richards, some years ago,
devised thirteen new forms of quotation punctemes to provide more
specificity to the kind of quotation a writer meant (Richards, 1974:xii-xiv).
The study of linguistics as well should begin including information
about punctuation in texts. As it is now, writing and punctuation doesn’t
205
even seem to exist in them, even in texts on syntax, the place where much of
the noticeable and most important punctuation occurs.
5.17 Teaching Punctuation
Judging from what is in the textbooks, articles, and books on teaching
punctuation, there is little wonder that students do not comprehend how to
do it. Teachers who use speech metaphors to teach writing are not doing their
students any favors. They are teaching how to drive a car by using
instructions for riding a horse. Speech is an oral and organically learned
technology that people acquire over a period of years by imitation. Writing
and punctuating are visual practices, a technology that is learned later by
translating phonemic information into graphemic information and then
practicing how to organize it so it is understandable to others who use the
same system. It is a different technology.
Teaching punctuation is a problem that has inspired many articles by
teachers in the past century. Just about every approach seems to have been
tried, most without much success. Some few have worked. John C. Shafer
made a good point in his article Punctuation and Process: A Matter of
Emphasis. He suggests, after some instruction in punctuation, handing out or
showing badly punctuated examples and asking the students to correct them
(Schafer, 1988:47). Often, when you see someone else’s errors, you start to
206
analyze the reason for the error, the result being your own avoidance of that
error.
Judith E. Moy, in Punctuating Punctuation in the English Classroom,
also proposed using Lewis Carroll to get students started. Hand out some of
his lines to students and let them put in the punctuation:
alice found a book which seemed to be in a peculiar
language but when she held it to the looking glass
she found that she could read this poem it seems very
pretty she said when she had finished it but its rather
hard to understand somehow it seems to fill my head
with ideas only i dont know exactly what they are
however somebody killed something thats clear an any rate
Students who have to complete the punctuation of this text will have to deal
with capitalization, apostrophes, terminal punctuation, quotation punctemes,
and a variety of comma locations—a significant challenge and an
entertaining lesson (Moy, 1996:50-60).
A study by Robert Stanley Zais in 1963 of the methods used to teach
punctuation revealed that an approach that emphasized only the “principle”
behind the use of a puncteme did not always improve the use of punctuation
in a student’s writing, but it could be more effective if it were taught by
combining the principle with more applications of it (Zais, 1963:94). Had a
better understanding of how punctuation was to work as part of a coherent
207
linguistic system been added to this recipe, students might have done better.
The characteristics of an individual puncteme are only a small part of a
larger plan.
One of the most aggressive attempts to find a better way to teach
punctuation was carried out by a number of scholars and teachers in
England. The Punctuation Project was organized at Manchester
Metropolitan University, and there was an online site devoted to it for some
time operated by Nigel Hall. A book titled Learning About Punctuation on the
subject was edited by Nigel Hall and Anne Robinson, and the articles in it
aired many of the roadblocks to teaching punctuation. Hall placed the
question front and center:
Is punctuation based on grammar, intonation, or a somewhat
accidental history? Or, is it a combination of the three?
(Hall, 1996:9.)
Hall points out that one barrier to getting to know how to punctuate may be
related to children’s lack of knowledge of the sentence, which might be
corrected to some extent by giving them well-punctuated materials to read as
well as texts that they may insert the punctuation in. “The expectation is
that children will develop a feel for what counts as a ‘sentence’” (Hall, 1996:
18).
208
One of the articles in the book, Conversations with Teachers About
Punctuation, by Anne Robinson, revealed that one of the main difficulties in
teaching punctuation to children indeed was:
working out what a sentence was. Notions of full stops and
capital letters make little sense without some concept of
sentence; after all, for the most part, what they mark are the
boundaries of sentences. Teachers were acutely aware that what
often appeared to many people outside education to be the most
basic and easiest thing of all, was actually one of the most
complex . . . .
(Robinson, 1996:82.)
Robert J. Scholes and Brenda J. Willis attempted to locate a
compromise between the two dominant approaches to punctuation: the
elocutionary function and the syntactic function. Some writers maintain that
punctuation is just a mixture of the two. They ultimately decided that “While
there may be cases in which punctuation corresponds to both prosodic and
syntactic information, the two are fundamentally incompatible” (Scholes and
Willis, 1990:14).
Nancy Mann labeled her lengthy article Point Counterpoint: Teaching
Punctuation As Information Management, perhaps partly because she had
trouble deciding which side was ultimately going to win, the prosodyists or
the grammarians. To conclude, she (perhaps not so jokingly) brings up the
idea of “style,” leaving us to decide ourselves who won the argument.
209
5.18 A Teaching Method for Punctuation
Although many people try to avoid it, there is ultimately no substitute for
learning at least a minimum of grammatical structures of the sort that are
most commonly encountered in writing and punctuating. One can learn to
write by “feel” over a period of time but will never arrive at a complete and
certain accuracy doing it that way. You can memorize thirty rules for
inserting commas, but if you can’t find those places because you don’t know
the grammar, the rules will do no good.
In order to punctuate with understanding, you need to see the sentence
as a form, a structure involving a subject and verb at least, in that order, and
understand that the role of punctuation is to protect it from becoming
cluttered with other words and phrases, realizing that the chief puncteme for
accomplishing this is the comma.
The use of the canonical sentence will substitute one whole structure
for the many rules, but, depending on the level of the students one is
teaching, it may still be necessary to teach some grammar. Teach the
sentence first, the schema, not the noun phrase and verb phrase. Slobin and
Bever confirmed that the canonical sentence has a presence in our language
behavior. Their abstract reads:
We propose that children construct a canonical sentence schema
as a preliminary organizing structure for language behavior.
The canonical sentence embodies the typical features of
210
complete clauses in the input language, and serves as a
framework for the application of productive and perceptual
strategies.
(Slobin and Bever, 1982:229.)
Their research was cross-linguistic with Turkish. They also noted other,
similar studies in which it was established that “children extract the basic
SVO order of the language” (Slobin and Bever, 1982:231). Since children have
this knowledge, we should certainly be using it to teach punctuation.
You could start with Noam Chomsky:
Colorless green ideas sleep furiously.
Most people will be able to identify the sentence here, despite its lack of
meaning, but they should look at some of the small parts as well. It makes
the point that there is a form that the sentence follows even if it can’t be
seen, and that is the form that is used in punctuation. Then notice something
besides the content words, some small parts, and move on to Lewis Carroll:
‘Twas brillig, and the slithy toves,
Did gyre and gimble in the wabe;
All mimsy were the borogoves,
And the mome raths outgrabe.
Townsend and Bever provided the evidence for the importance of function
words in the comprehension of language: research has shown that function
words are noticed more quickly than content words (Townsend and Bever,
2001:188). Function words are, after all, the skeleton of the language, words
211
you see more often than any content words, and the clues to capturing the
outline of a sentence. You could teach one hundred minutes of grammar,
mainly function words and basic structures, so students get at least a
minimal grasp of the sentence and the way it can be manipulated and held
together with function words. One could also notice the “reputable writers
approach” in the way Carroll punctuated the lines.
Study the two axes of the sentence, the static vertical axis as well as
the dynamic horizontal axis, and realize the differences between the
punctemes in those two locations their different chores. Then study the
individual punctemes and point out the special and dual use of each. This will
create some order in what appears at first to be a random collection of
unrelated symbols. Teach the sentence as a graphic with the graphemes in
boxes.
Sentence combining is also a good way to get students to see the
possibilities for manipulating sentence parts and start seeing the junction
points that may need some punctuation. Studying and writing different kinds
of sentences, such as cumulative sentences and periodic sentences, will also
make students study punctuation more because it will be more necessary in
those kinds of sentences.
Should students become adept at using punctuation, writing
interesting and stylish sentences could easily become a form of entertainment
212
for them, a challenge to see who could write the most interesting one.
Education, after all, should not be all drudgery.
Have students write a short sentence, like the one you just read, a
sentence which has many parts to it, a sentence which moves along from one
phrase to another, adding ideas and structures that elaborate on something
that they are interested in, making it perhaps a useful sentence for
describing some complicated or detailed thing, such as a circus or the parade
they last saw, a cluttered room or some other detail-filled scene, aiming
always at quality instead of quantity, but keeping in mind that so much of it
depends on the use of the function words and punctuation to make it work so
that a reader will not have trouble with its construction and be able to enjoy
a cumulative sentence.
The idea of the cumulative sentence was made popular by Francis
Christensen and others in the 1960s, but it could be that it was already in the
thoughts of a medieval writer, Geoffrey of Vinsauf, whose discussion of
“amplification” of a sentence using description, circumlocution, digression,
personification, and apostrophe—and “a single verb is sufficient for extending
material endlessly”—could very well have been the first person to work on
the idea (Vinsauf, 1968:57).
Good periodic sentences, on the other hand, which involve thinking
ahead instead of backwards, with a full understanding of how to use
punctuation, by someone who has gotten good practice at writing longer
213
sentences, in the same way as an athlete who gets out to practice every day,
regardless of other circumstances, which do sometimes curtail one’s ability to
perform, are harder to write.
5.19 Summary
The new paradigm for punctuation has many things to offer for a better
understanding of how individual punctemes function. But it is necessary to
see them as visual symbols operating in a graphic structure. The sentence
form is the place where they perform, using the system model of the
canonical sentence.
Research on punctuation can be made more effective because the use of
a stable schema will allow for better comparisons between different varieties
of English as well as the stylistic behavior of individual writers.
Having a clearer understanding of how punctuation works should
make teaching it easier because the categories of punctemes can be visually
located and classified rather than be dealt with one grammatical location at a
time in the way they are in many handbooks.
214
Chapter 6
Postscript
6.1 Some Things Don’t Change and Some Might
Having done some significant work in researching and writing this study, I
would like to think that it will be accepted as a useful tool for teaching
punctuation and will replace all of the elocutionary-related schemes that
have been around for quite some time. I suspect that I will be mistaken about
that because there are some advocates for using speech as an avenue for
understanding punctuation that have published their views and are willing
to express their beliefs that the comparison to speech is the only way. We will
look at some of their views.
6.2 Punctuation: Sight or sound?
Yes, it may be time to admit that there are still some scholars who believe
without question that the use of speech to help determine where to punctuate
is really the only way. John Dawkins is one of them. He has published his
views frequently. His approach is to establish a hierarchy of features that he
has selected from “good writers of nonfiction,” probably using the “reputable
writers approach” [ . ; : -- , 0 ]. “ The secret . . . is for student writers to do
what good writers of nonfiction do,” which is to use meaning not grammarbased rules” for punctuating (Dawkins, 2003:155-156). Apparently, the
students are to read their sentences out loud and determine which feature of
215
punctuation would be most appropriate at a particular junction in a sentence
based on the intended meaning. He concludes this way:
My study of punctuation suggests that the purpose of all six of
the hierarchical marks is rhetorical. The marks were
undoubtedly created for rhetorical purposes, that is, to make
meaning more clear, more effective; in other words, the concern
was with meaning . . . .
(Hawkins, 2003:161.)
Peter Elbow, in his book Vernacular Eloquence: What Speech Can
Bring to Writing, rather than having a list of features or a hierarchy, just
believes, pragmatically, that most people don’t know the punctuation “rules”
well enough to do a good job of punctuating. Instead, we should read our
writing out loud after some training to develop a technique called “careful
reading” (Elbow, 2012:281). Elbow does provide a list of five rules of his own
for “Punctuating Defensively” using his “read aloud” approach (Elbow, 2012:
283-285). He is certainly correct in observing that writers are seldom told
when they are punctuating properly, only when they do it wrong.
6.3 The Cross-Linguistic Study of Punctuation
My original intent in studying punctuation was to do a cross-linguistic study.
It was with wild and uncontrolled optimism that I began to think about such
a study of punctuation, the intent being to look at every written language in
the world and see what they were doing in the way of punctuation. The
216
earliest work in that direction ran into barriers quickly. In the first place, it
is very difficult to even determine the number of written languages. Besides,
languages come and go. Some have no writing systems but are working on
developing them. But the problems of looking at too many languages was
insurmountable. There may or may not be time to accomplish some part of
that, but this examination of English punctuation had to come first.
In addition to dealing with things numerically, there was the question
of how punctuation actually works. We know that languages other than
English have a different word order so studying their punctuation may
require a stable standard for comparison but based on a different canonical
form. Then there are the dramatic differences between word order and
inflectional languages.
We can imagine that using the “grammar or prosody” approach would
probably not work well in analyzing most languages. Using a linguistic
approach might provide a uniform standard that would allow a fair
comparison. Languages that pattern closer to English might be a good place
to start just to see how complicated the study might be. We might wonder, for
example, just how different World Englishes are from one another in their
use of punctuation. Do they tend to subscribe to either British or American
style, or do they adopt one of their own?
Greta Little, some years ago, was eager to see some cross-linguistic
studies done:
217
From crosslinguistic and diachronic studies we will be able to
understand better what universal principles exist for language
in the written mode. . . . Attempting to understand the function
of punctuation is crucial to our study of the linguistic and
cognitive parameters of literacy.
(Little, 1986:84-85.)
To accomplish her goal, however, a clear and more standard approach to
punctuation is necessary, one that can be applied universally. Adopting
a language’s word order schema might be a workable start.
6.4 Conclusion
This study was certainly not intended to support the “rules” that are widely
published as a way to participate in using punctuation. The intent instead
was to reveal another way of looking at writing as a graphic system that
displays the canonical sentence using gestalt visual techniques. To
understand this approach requires putting aside the rhetorical/prosodic/
elocutionary and grammatical approaches and accepting written English and
its sentences and partials as a visual representation of the language which
has its own linguistic way of displaying the punctuation. Only time will tell if
this paradigm finds any acceptance. However, it should be seen as an
invitation for more research rather than a complete solution. As Thomas
Kuhn said it, “no paradigm that provides a basis for scientific research ever
completely resolves all its problems” (Kuhn, 1970:79).
218
References
Altman, Gabriel. 2008. Toward a theory of script. Analyses of script:
properties of characters and writing systems, ed. by Gabriel Altman
and Fan Fengxiang, 149-164. Berlin: Mouton de Gruyter.
Altman, Rochelle. 2004. Absent voices: the story of writing systems in the
west. New Castle, DE: Oak Knoll Press.
Andrews, Edna. 1990. Markedness theory: The union of asymmetry and
semiosis language. Durham, NC: Duke University Press.
Aronoff, Mark. 1994. Spelling as culture. Writing systems and cognition, ed.
by W. C. Watt, 67-86. Dordrecht: Kluwer Academic Publishers.
Arriola-Nickell, Gail Emily. 2003. Punctuation as symbol: experiencing
archetypal patterns through personal narrative. Santa Barbara, CA:
Pacific Graduate Institute dissertation.
Arn, Mary-Jo. 1994. On punctuating medical literary texts. Text 7. 161-174.
Arnheim, Rudolf. 1969. Visual thinking. Berkeley, CA: University of
California Press.
Baldwin, Scott R. and James M. Coady. 1978. Psycholinguistic approaches to
a theory of punctuation. Journal of Literacy Research 10.363-375.
Bally, Charles; Albert Sechehaye; and Albert Riedlinger (eds.) 1959. Course
in general linguistics Ferdinand de Saussure. New York: McGraw-Hill
Book Co.
Baron, Naomi. 2001. Commas and canaries. Language Science 23.15-67.
Bartlett, Frederic C. 1995. Remembering: a study in experimental and social
psychology. Cambridge, MA: Cambridge University Press.
Battistella, Edwin. 1993. Notes on the sign structure of English punctuation.
The Peirce seminar papers: An annual of semiotic analysis, vol. 1, ed.
by Michael Shapiro. 7-30. Providence, RI: Berg Publishers.
Battistella, Edwin. 1996. Marked and unmarked punctuation signs in
English. Prague linguistic circle papers, vol. 2, ed. by Eva Hajicova;
219
Oldrich Leska; Petr Sgall; and Zdena Skoumalova, 235-248.
Amsterdam: John Benjamins.
Battistella, Edwin. 1999. A Peircean approach to the apostrophe. The Peirce
seminar papers: essays in semiotic analysis, Vol. 4. Proceeding of the
International Colloquium on Language and Peirce Sign Theory. Duke
University, June 19-21, ed. by Michael Shapiro, 87-119. New York:
Berghahn Books.
Bayer, Herbert. 2009. On typography, 1967. Graphic design theory: readings
from the field, ed. by Helen Armstrong. New York: Princeton
Architechtural Press
Beckett, Dave. 1994. Biography: William of Ockham. Principia Cybernetica
Web (Principia Cybernetica, Brussels), ed. by F. Heylighen; C. Joslyn;
and V. Turchin. URL http:// cleamc11.vub.ac.be/Occam’s Razor.
Behrens, Roy R. 1998. Art, design and gestalt theory. Leonardos
31.299-303.
Bernhardt, Stephen A. 1993. The shape of text to come: the texture of print
on screens. College Composition and Communication 44.151-175.
Bever, Thomas G.; Steven Jandreau; Rebecca Burwell; Ron Kaplan; and
Annie Zalnen. 1990. Spacing printed texts to isolate major phrases
improves readability. Visible Language 25.74-87.
Bloomfield, Leonard. 1933. Language. New York: Holt, Rinehart and
Winston.
Bolton, W. F. 1982. A living language: the history and structure of English.
New York: Random House.
Brendel, Elke, Jörg Meibauer, and Markus Steinback (eds.) 2001.
Understanding quotation. Boston, MA: De Gruyter.
Bringhurst, Robert. 2005. The elements of typographic style. Point Roberts,
WA: Hartley and Marks Publishers.
Brown, Kieth, and Jim Miller. 1996. Concise Encyclopedia of Syntactic
Theories. Cambridge, MA: Cambridge University Press.
220
Bruthiaux, Paul. 1995. The rise and fall of the semicolon: English
punctuation theory and English teaching practice. Applied linguistics
16.1-14.
Carroll, Lewis. 1999. Through the looking glass, ed. by Susan L. Rattiner.
Mineola, NY: Dover Publications.
Chafe, Wallace. 1988. Punctuation and the prosody of written language.
Written Communication 5.395-426.
Chang, Demmpsey; Campbell Wilson; and Laurence Dooley. 2003-2004.
Towards criteria for visual layout of instructional multimedia
interfaces. Journal of Educational Technology Systems 32.3-29.
The Chicago manual of style 16th ed. 2010. Chicago, IL: University of
Chicago Press.
Childs, Edmund. 1976. William Caxton: a portrait in a background. New
York: St. Martin’s Press.
Chomsky, Noam. 1966. Syntactic structures. The Hague: Mouton & Co.
Coe, Michael D. 1992. Breaking the Maya code. New York: Thames and
Hudson.
Clanchy, M. J. 1993. From memory to written record: England 1066-1307. 2nd
ed. Oxford, UK: Blackwell.
Clark, Romy, and Roz Ivanic. 1997. The politics of writing. London:
Routledge.
The compact edition of the Oxford English dictionary. 1971. New York:
Oxford University Press.
Connors, Robert J, and Andrea A. Lunsford. 1988. Frequency of formal errors
in current college writing, or ma and pa kettle do research. College
Composition and Communication 39.395-409.
Coulmas, Florian. 1991. The writing systems of the world. Oxford: Blackwell.
Coulmas, Florian. 1999. The Blackwell encyclopedia of writing systems.
Oxford: Blackwell.
221
Coulmas, Florian. 2003. Writing systems: an introduction to their linguistic
analysis. Cambridge: Cambridge University Press.
Cram, David. 1989. Seventeenth-century punctuation theory: Butler’s
philosophical analysis and Wilkins’ philosophical critique. Folio
Linguistica Historica 7.309-349.
Crystal, David. 1995. The Cambridge encyclopedia of the English language.
Cambridge: Cambridge University Press.
Crystal, David. 1997. A dictionary of linguistics and phonetics. 134-135.
Oxford: Blackwell.
Daelemans, Walter, Koenraad De Smed, and Gerald Gazdar. 1992.
Inheritance in natural language processing. Computational Linguistics
18.205-218.
Daniels, Peter T. 1992. Is a structured graphemics possible? LACUS Forum
18.528-537.
Daniels, Peter T. 1996. The study of writing systems. The world’s writing
systems, ed. by Peter T. Daniels and William Bright, 3-17. New York:
Oxford University Press.
Daniels, Peter T. 2001. Writing systems. The handbook of linguistics, ed. by
Mark Aronoff and Janie Rees-Miller, 43-80. Oxford: Blackwell.
Daniels, Peter T., and William Bright (eds.) 1986. The world’s writing
systems. New York: Oxford University Press.
Dawkins, John. 2003. Teaching meaning-based punctuation. Teaching
English in the two-year college 31.154-62.
Deane, Paul D. 1991. Syntax and the brain: neurological evidence for the
spatialization of form hypothesis. Cognitive Linguistics 2.361-367.
DeFrancis, John. 1989. Visible speech: the diverse oneness of writing systems.
Honolulu: University of Hawaii Press.
Deneau, Daniel. P. 2003. Reading Simon’s (parentheses) The Flanders Road.
Neophilologus 87.553-561.
222
DeVinne, Theodore Low. 1902. Correct composition: a treatise on spelling
abbreviations, the compounding and division of words, the proper use
of figures and numerals, italic and capital letters, notes, etc. with
observations on punctuation and proof-reading. New York: The
Century Co.
Dillon, J. T. 1982. In pursuit of the colon: a century of scholarly progress:
1880-1980. Journal of Higher Education 51.93-99.
Diringer, David, and David R. Olson. 1989. Alphabetic writing. Encyclopedia
Britannica, 15th ed 1040-1049. Chicago: Encyclopedia Britannica.
Dominowski, Roger L., and Pamela Dallob. 1995. Insight and problem
solving. The nature of insight, ed. by Robert J. Sternberg and Janet E.
Davidson, 33-62. Cambridge, MA: MIT Press.
Dondis, Donis. A. 1973. A Primer of visual literacy. Cambridge, MA: MIT
Press.
Doran, Christine D. 1998. Incorporating punctuation into the sentence
grammar: a lexicalized tree adjoining grammar perspective.
Philadelphia, PA: University of Pennsylvania dissertation.
Downing, John, and Che Kan Leong. 1982. Psychology of reading. New York:
Macmillan.
Elbow, Peter. 2012. Vernacular Eloquence: what speech can bring to writing.
New York: Oxford University Press.
Eldredge, Niles, and Stephan Jay Gould. 1972. Punctuated equilibria: an
alternative to phyletic gradualism. Models in Paleobiology, ed. by
Thomas J. M. Schopf, 81-115. San Francisco, CA: Freeman, Cooper and
Company.
Ellsworth, Blanche, and John A. Higgins. 1985. English simplified, 5th ed.
New York: Harper & Row.
Esser, Jürgen. 2000. Medium-transferability and presentation structures in
speech and writing. Journal of Pragmatics 32.1523-1538.
223
Fodor, J. A.; T. B. Bever; and M. F. Garrett. 1974. The psychology of
language: An introduction to psycholinguistics and generative
grammar. New York: McGraw-Hill.
Foster, Donald W. 1989. Elegy by w.s.: a study in attribution. Newark, NJ:
University of Delaware Press.
Foster, Donald W. 2000. Author unknown: on the trail of anonymous. New
York: Henry Holt and Company.
Francis, W. Nelson. 1958. The structure of American English. New York: The
Ronald Press.
Frawley, William J. 2003. International encyclopedia of linguistics. Oxford:
Oxford University Press.
Frazier, Lyn. 1979. On comprehending sentences: Syntactic parsing
strategies. University of Connecticut dissertation. Indiana University
Linguistics Club.
Frazier, L., and J. D. Fodor. 1978. The sausage machine: a new two-stage
parsing model. Cognition 6.291-325.
Frazier, Lyn, and Charles Clifton, Jr. 1996. Construal. Cambridge, MA: A
Bradford Book.
Garber, Marjorie. 2003. Quotation marks. New York: Routledge.
Gelb, I. J. 1965. A study of writing. Rev. ed. Chicago: University of Chicago
Press.
Gerbner, George. 1960. The interaction model: Perception and
communication. Research, principles, and practices in visual
communicatio, 4-15. Washington, DC: The Association for Educational
Communication and Technology.
Giardino, Valeria. 2013. Towards a diagrammatic classification. The
Knowledge Engineering Review 28.237-248.
Givon, T. 1990. Syntax: A functional-typological introduction, vol. 2.
Amsterdam: John Benjamins.
224
Gorrell, Robert M., and Charlton Laird. 1967. Modern English handbook.
Englewood Cliffs, NJ: Prentice-Hall.
Gould, Stephan Jay. 2007. Punctuated equilibrium. Cambridge, MA. Harvard
University Press.
Graham-White, Anthony. 1995. Punctuation and its dramatic value in
Shakespearean drama. Newark, NJ: University of Delaware Press.
Greenbaum, Sidney. 1996. The Oxford English grammar. New York: Oxford
University Press.
Grindlay, Benjamin. 2002. Missing the point: the effect of punctuation on
reading performance. Adelaide, Australia: Adelaide University
dissertation.
Hall, Nigel. 1996. Learning about punctuation: an introduction and overview.
Learning about punctuation, ed. by Nigel Hall and Anne Robinson, 536.Portsmouth, NH: Heinemann.
Hall, Nigel, and Anne Robinson (eds.) 1996. Learning about punctuation,
Portsmouth, NH: Heinemann.
Halliday, Michael 1967. Notes on transivity and theme in English, part II.
Journal of Linguistics 3.199-244.
Halliday, M. A. K. 1989. Spoken and written language. Oxford: Oxford
University Press.
Hannas, William C. 1997. Asia’s orthographic dilemma. Honolulu, HI:
University of Hawai’i Press.
Harris, Roy. 1995. Signs of writing. New York: Routledge.
Harris, Zellig. 1982. A grammar of English on mathematical principles. New
York: John Wiley & Sons.
Henderson, Leslie. 1985. On the use of the term “grapheme.” Language and
Cognition Processes 1.135-148.
Hill, Robin. 1996. A comma in parsing: a study into the influence of
punctuation (commas) on contextually isolated “garden-path”
sentences. Dundee UK: University of Dundee dissertation.
225
Honan, Park. 1960. Eighteenth and nineteenth century punctuation theory.
English Studies 41.92-102.
Husband, T. F., and M. F. A. Husband. 1905. Punctuation: its principles and
practice. London: George Routledge & Sons Limited.
Iran-Nejad, Asghar. 1980. The schema: a structural or a functional pattern.
Technical report no. 159. Urbana-Champaign IL: University of Illinois.
Jacobs, David W. 1988. The use of grouping in visual object recognition.
Cambridge, MA: MIT masters thesis.
Johnson-Laird, Philip N. 1983. Mental models: towards a cognitive science of
language, inference, and consciousness. Cambridge, MA: Harvard
University Press.
Johnson-Sheehan, Richard. 2005. Technical communication today. New York:
Pearson Longman.
Jones, Bernard. 1996. What’s the point? A (computational) theory of
punctuation. Edinburgh, UK: University of Edinburgh dissertation.
Kanizsa, Gaetano. 1979. Organization in vision: Essays on gestalt perception.
New York: Praeger Scientific.
Kaye, Alan S. 1988. English does have diacritics. Language Monthly 57.11.
King, Martin Luther, Jr. 1963. Letter from Birmingham jail. The Atlantic
Monthly 212.78-88.
Koffi, Ettien. 1995. Indigenizing punctuation marks. Notes on Literacy
21.1-11.
Knobler, Nathan. 1966. The visual dialogue: An introduction to the
appreciation of art. New York: Holt, Rinehart and Winston.
Kuhn, Thomas S. 1970. The structure of scientific revolutions. Chicago:
University of Chicago Press.
Kuiper, Shirley, and Cheryl M. Luke. 1992. Report writing with
microcomputer applications. Cincinnati: OH South-western
Publishing Co.
226
Kurzon, Dennis. 2008. A brief note on diacritics. Written Language and
Literacy 11.90-94.
Kuteva, Tania. 1999. Languages and societies: the “punctuated equilibrium”
model of language development. Language and Communication 19.
213-228.
Lakoff, George. 1987. Women, fire, and dangerous things: what categories
reveal about the mind. Chicago: University of Chicago Press.
Lappin, Shalom. 2001. An introduction to formal semantics. The handbook of
linguistics ed. by Mark Aranoff and Janie Rees-Miller, 369-393.
Malden, MA: Blackwell Publishers.
Lauchman, Richard. 2010. Punctuation at work: simple principles for
achieving clarity and good style. New York: American Management
Association.
Lennard, John. 1991. But I digress: the exploitation of parentheses in English
`
printed verse. Oxford: Clarendon Press.
Lennard, John. 2011. In/visible punctuation. Visible Language 45.121-138.
Levinson, Joan Persily. 1985. Punctuation and the orthographic sentence:
a linguistic analysis. New York: City University of New York
dissertation.
Levinson, Joan Persily. 1989. The linguistic status of the orthographic (text)
sentence. CUNY Forum 14.113-117.
Limaye, Mohan R. 1983. Approaching punctuation as a system. The ABCA
Bulletin 46.28-33.
Little, Greta D. 1984. Punctuation. Research in composition and rhetoric: a `
bibliographic source book, ed. by Michael G. Moran and Ronald F.
Lunsford, 371-398. West Port, CN: Greenwood Press.
Little, Greta. 1986. Punctuation as a linguistic phenomenon: research issues.
SECOL Review 10.69-91.
Lockwood, David G. 2001. Phoneme and grapheme: how parallel are they?
LACUS Forum 27.307-316.
227
Long, Ralph Bernard. 1961. The sentence and its parts: a grammar of
contemporary English. Chicago: University of Chicago Press.
Lounsbury, Floyd G. 1989. The ancient writing of middle America. The
origins of writing, ed. by Wayne M. Senner, 203-337. Lincoln, NB:
University of Nebraska Press.
Mandell, Alan. 1974. The language of science. Washington, DC: National
Science Teachers Association.
Mann, Nancy. 2003. Point counterpoint: teaching punctuation as information
management. College Composition and Communication. 54.359-393.
Mayer, Richard E. 1995. The search for insight: Grappling with gestalt
psychology’s unanswered questions. The nature of insight, ed. by
Robert J. Sternberg and Janet E. Davidson, 3-32. Cambridge, MA: MIT
Press.
Meyer, Charles F. 1983. A descriptive study of American punctuation.
Milwaukee WI: University of Wisconsin-Milwaukee dissertation.
Meyer, Charles F. 1987. A linguistic study of American punctuation. New
York: Peter Lang.
Meyer, Leonard B. 1958. Emotion and meaning in music. Chicago: University
of Chicago Press.
Moore, Colette. 2011. Quoting speech in early English. Cambridge, MA:
Cambridge University Press.
Moravcsik, Edith. 2006. An introduction to syntax: fundamentals of syntactic
analysis. New York: Continuum.
Moravcsik, Edith, and Jessica Wirth. Markedness—an overview. 1986.
Markedness, ed. by Fred R. Eckman; Edith Moravcsik; and Jessica
Wirth, 1-11. New York: Plenum Press.
Mountford, John. 1980. Writing-system as a concept in linguistics.
Information Design Journal 1.223-231.
Moy, Judith. 1996. Punctuating punctuation in the English classroom.
English Teaching Forum 34.50-60.
228
Mulderig, Gerald P., and Landon Elsbree. 1990. The Heath handbook.
Lexington, MA: D. C. Heath and Co.
Murray, Lindley.1816. An English grammar comprehending the principles
and rules of the language illustrated by appropriate exercises and a key
to the exercises. York, UK: Thomas Wilson & Sons. High Ausgate.
Nunberg, Geoffrey. 1990. The linguistics of punctuation. Stanford CA: Center
for the Study of Language and Information.
Nunberg, Geoffrey. 1998. About semicolons they were never wrong, the old
masters. Paper delivered at a Modern Language Association
Convention session. San Francisco, CA.
Nunberg, Geoffrey, Ten Briscoe, and Rodney Huddleston. 2002. Punctuation.
The Cambridge grammar of the English language, ed. by Rodney
Huddleston, Geoffrey K. Pullum, et al., 1723-1764. Cambridge, MA:
Cambridge University Press.
Olson, David. 1977. From utterance to text: the bias of language in speech
and writing. Harvard Educational Review 47. 257-281.
Ong, Walter J. 1944. Historical backgrounds of Elizabethan and Jacobean
punctuation theory. PMLA 59.349-360.
Ong, Walter J. 1982. Orality and literacy: the technologizing of the word. New
York: Methuen.
Ong, Walter J. 1983. Ramus: method, and the decay of dialogue: from the art
of discourse to the art of reason. Cambridge: Harvard University Press.
Ong, Walter J. 1992. Writing is a technology that restructures thought. The
linguistics of literacy, ed. by Pamela Downing; Susan D. Lima; and
Michael Noonan, 293-319. Amsterdam: John Benjamins.
The Oxford American college dictionary. 2002. New York: G. P. Putnam’s
Sons.
Parkes, M. B. 1978. Punctuation, or pause and effect. Medieval eloquence:
studies in the theory and practice of medieval rhetoric, ed. by James J.
Murphy, 127-142. Berkeley, CA: University of California Press.
229
Parkes, M. B. 1993. Pause and effect: an introduction to the history of
punctuation in the west. Berkeley: University of California Press.
Partridge, A. C. 1964. Orthography in Shakespearean and Elizabethan
drama: a study of colloquial contractions, elision, prosody and
punctuation. Lincoln, NE: University of Nebraska Press.
Partridge, Eric. 1953. You have a point there: a guide to punctuation and its
allies. London: Routledge & Kegan Paul.
Patt, Sebastian. 2013. Punctuation as a means of medium-dependent
presentation structure in English. Tübingen: Narr Francke Attempto
Verlag.
Petit, Angela. 2003. The stylish semicolon: teaching punctuation as rhetorical
choice. English Journal 92.66-72.
Poyatos, Fernando. 1981. Punctuation as nonverbal communication: toward
an interdisciplinary approach to writing. Semiotica 34.91-112.
Pulgram, Ernst. 1951. Phoneme and grapheme: A parallel. Word 7.15-20.
Quirk, Randolph; Sidney Greenbaum; Geoffrey Leech; and Jan Svartvik.
1985. A comprehensive grammar of the English language. New York:
Longman.
Richards, I. A. 1974. Beyond. New York: Harcourt Brace Jovanovich.
Robinson, Anne. 1996. Conversations with teachers about punctuation.
Learning about punctuation, ed. by Nigel Hall and Anne Robinson, 7491. Portsmouth, NH: Heinemann.
Robinson, Ian. 1992. Medieval punctuation, the concept of the sentence and
reading. Medieval History. 2.36-48.
Rogers, Henry. 2005. Writing systems: a linguistic approach. Oxford:
Blackwell.
Rousseau, S. 1813. Punctuation: on an attempt to facilitate the art of pointing,
on the principle of grammar and reason. London: Longman, Hurot,
Rees, Orme & Brown.
230
Rumelhart, David E. 1980. Schemata: the building blocks of cognition.
Theoretical issues in reading comprehension, ed. by Rand J. Spiro,
Betram C. Bruce, and William F. Brewer, 33-58. Hillsdale, NJ:
Laurence Erlbaum Associates.
Saenger, Paul. 1982. Silent reading: its impact on late medieval script and
society. Viator: Medieval and Renaissance Studies 13.367-414.
Saenger, Paul. 1997. Space between words: the origins of silent reading.
Stanford, CA: Stanford University Press.
Saenger, Paul, and Michael Heinlen. 1991. Incunable description and its
implication for the analysis of fifteenth-century reading habits.
Printing the written word: the social history of books, 1450-1520, ed. by
Sandra Hindman, 225-258. Ithica, NY: Cornell University Press.
Salmon, Vivian. 1962. Early seventeenth-century punctuation as a guide to `
sentence structure. Review of English studies, new series. 13.347-360.
Salmon, Vivian. 1979. The study of language in 17th-century England.
Amsterdam: John Benjamins.
Salmon, Vivian. 1988. English punctuation theory 1500-1800. Anglia 106.
285-314.
Salmon, Vivian. 1999. Orthography and punctuation. The Cambridge history
of the English language, III, 1476-1776, ed. by Roger Lass, 13-55.
Cambridge: Cambridge University Press.
Sampson,, Geoffrey. 1985. Writing systems: a linguistic introduction.
Stanford, CA: Stanford University Press.
Sapir, Edward. 1921. Language. New York: Harcourt, Brace & World.
Say, Bilge. 1998. An information-based approach to punctuation. Ankara
Turkey: Bilkent University dissertation.
Schafer, John C. 1988. Punctuation and process: a matter of emphasis.
English Journal 77.46-49.
Shaughnessy, Mina P. 1977. Errors & expectations: a guide for the teacher of
basic writing. New York: Oxford University Press.
231
Scholes, Robert James. 1999. A linguistic approach to reading and writing.
Lewiston, ME: The Edwin Mellen Press.
Scholes, Robert J., and Brenda J. Willis. 1990. Prosodic and syntactic
functions of punctuation—a contribution to the study of orality and
literacy. Interchange 21.13-20.
Schou, Karsten. 2002. The problem of English punctuation. Roskilde,
Denmark: Roskilde University dissertation.
Schou, Karsten. 2007. The syntactic status of English punctuation. English
Studies 88.195-216.
Shieber, Stuart M., Xiaopeng Tao. 2003. Comma restoration using
constituency information. Proceedings of HLT-NAACL, Main Papers,
Edmonton. 142-148.
Shimojima, Atsushi. 2001. The graphic-linguistic distinction. Artificial
Intelligence Review. 15.5-27.
Simpson, Percy. 1911. Shakespearian punctuation. Oxford: Clarendon Press.
Sklar, Elizabeth S. 1976. The progressive apostrophe: the development and
decline of a crooked mark. College English 38.175-183.
Slobin, Dan I., and Thomas G. Bever. 1982. Children use canonical sentence
schema: a crosslinguistic study of word order and inflections. Cognition
12.229-265.
Smith, P. T.; T. R. Meredith; H. M. Pattison; and C. M. Sterling. 1984. The
representation of internal word structure in English. Orthographies
and reading, ed.by Leslie Henderson, 103-120. Hillsdale, NJ: Laurence
Erlbaum.
Solomon, Martin. 2004. The power of punctuation. Visual rhetoric in a digital
world: A critical sourcebook, ed. by Carolyn Handa, 282-289. Boston,
MA: Bedford/St.Martins
Southhall, Richard. 1984. First principles of typographic design for document
production. TUGboat 5.79-90.
232
Steedman, Mark. 2000. Information structure and the syntax-phonology
interface. Linguistic Inquiry 31.649-689.
Steegar, David. 1975. Prosody and punctuation: a linguistic and experimental
study. Toronto, Canada: University of Toronto dissertation.
Stevick, Robert D. 2004. Graphotactics of the Old English ‘Alexander’s Letter
to Aristotle.’ Studia Anglica Posnaniensia 40.3-13.
Stolz, Walter Sargent. 1964. Syntactic constraint in spoken and written
English. Madison, WI: University of Wisconsin dissertation.
Summey, George. 1919. Modern punctuation: its utilities and conventions.
New York: Oxford University Press.
Summey, George. 1949. American punctuation. New York: The Ronald Press
Co.
Thompson, Edward Maunde. 1966. An introduction to Greek and Latin
paleography. New York: Burt Franklin.
Townsend, David J., and Thomas G. Bever. 2001. Sentence comprehension:
the integration of habits and rules. Cambridge MA: MIT Press.
Trask, R. L. 1997. A student’s dictionary of language and linguistics. New
York: Arnold.
Trask, R. L. 1999. Key concepts in language and linguistics. New York:
Routledge.
Treip, Mindele. 1970. Milton’s punctuation: and changing English usage,
1582-1676. New Fetter Lane, UK: Methuen & Co. Ltd.
Trigger, Bruce G. 2004. Writing systems: a case study in cultural evolution.
The first writing: script inventions as history and process, ed. by
Stephen D. Houston, 39-68. New York: Cambridge University Press.
Truss, Lynne. 2003. Eats, shoots & leaves: The zero tolerance approach to
punctuation. New York: Gotham Books.
Tufte, Edward R. 1990. Envisioning information. Cheshire, CN: Graphics
Press.
233
Tufte, Edward R. 1997. Visual explanations: Images and quantities, evidence
and narrative. Cheshire, CN: Graphics Press.
Tuite, Thomas. 1726. The Oxford spelling book: being a complete introduction
to English orthography. London: Joseph Hazard.
Turnbull, Arthur T., and Russell N. Baird. 1975. The graphics of
communication: Typography, layout, design, 3rd. ed. New York: Holt,
Rinehart and Winston.
Twyman, Michael. 1979. A schema for the study of graphic language.
Processing of visible language, vol. 1, ed. by Paul A. Kolers; Merald E.
Wrolstad; and Herman Bouma, 117-150. New York: Plenum Press.
Vachek, Josef. 1989. Written language revisited. Philadelphia, PA: John
Benjamins.
VanWinkle, C. S. 1818. The printers guide; or, an introduction to the art of
printing: including an essay on punctuation and remarks on
orthography. New York: C. S. VanWinkle.
Venezky, Richard L. 1970. The Structure of English Orthography. The Hague:
Mouton.
Vial, Stephanie D. 2000. Take pause: musical punctuation in the eighteenth
century. Ithaca, NY: Cornell dissertation.
Vinsauf, Geoffrey.1968. Instruction in the method and art of speaking and
versifying, trans. by Roger P. Paar. Milwaukee, WI: Marquette
University Press.
Waller, Robert H. W. 1980. Graphic aspects of complex texts: typography as
macro-punctuation. Processing of visible language, Vol. II, ed. by P. A.
Kolers, M. E. Wrolstad, and H. Bourna, 241-253. New York: Plenum
Press.
Waller, Robert H. W. 1982. Text as diagram: Using typography to improve
access and understanding. The technology of text, ed. by D. Jonassen,
137-166. Englewood Cliffs, NJ: Educational Technology Publications.
234
Waller, Robert. 1988. The typographic contribution to language: towards a
model of typographic genres and their underlying structures. Reading
Berkshire, UK: University of Reading dissertation.
Wells, J. C. 2000. Orthographic diacritics and multilingual computing.
Language Problems and Language Planning 24.249-272.
Wertheimer, Max. 2012. On perceiving motion and figural organization, ed.
by Lothar Spillmann. Cambridge, MA: MIT Press.
Whitehall, Harold. 1956. Structural essentials of English. New York:
Harcourt, Brace and Company.
Wilson, John. 1864. A treatise on English punctuation, 15th ed. Boston, MA:
Crosby and Nichols.
Wingo, Otha. 1972. Latin punctuation in the classical age. The Hague:
Mouton.
Yaggy, Elinor. 1953. Let’s take the guesswork out of punctuation. College
Composition and Communication 4.128-131.
Zais, Robert Staley. 1963. A study to determine whether a punctuation
symbol’s characteristics influence the effectiveness of a method of
teaching punctuation. Providence, RI: Brown University Master’s
thesis.
235
CURRICULUM VITAE
Albert E. Krahn
Place of birth: Milwaukee, Wisconsin
Education:
B.S., University of Wisconsin-Milwaukee, January 1963
Major: English
M.A., University of Wisconsin-Milwaukee, May 1973
Major: English
Teaching Experience:
Milwaukee Area Technical College, 1968 to 1999
Teaching English, Communication Skills, Technical
Communication, Modern Cinema, Communication
Skills Telecourse, Communication Skills Online.
Dissertation Title: A New Paradigm for Punctuation